Anda di halaman 1dari 70

Informatica PowerCenter (Version 9.5.

1)
Big Data Edition
Informatica PowerCenter Big Data Edition
Version 9.5.1
December 2012
Copyright (c) 2012 Informatica. All rights reserved.
This software and documentation contain proprietary information of Informatica Corporation and are provided under a license agreement containing restrictions on use and
disclosure and are also protected by copyright law. Reverse engineering of the software is prohibited. No part of this document may be reproduced or transmitted in any form,
by any means (electronic, photocopying, recording or otherwise) without prior consent of Informatica Corporation. This Software may be protected by U.S. and/or international
Patents and other Patents Pending.
Use, duplication, or disclosure of the Software by the U.S. Government is subject to the restrictions set forth in the applicable software license agreement and as provided in
DFARS 227.7202-1(a) and 227.7702-3(a) (1995), DFARS 252.227-7013

(1)(ii) (OCT 1988), FAR 12.212(a) (1995), FAR 52.227-19, or FAR 52.227-14 (ALT III), as applicable.
The information in this product or documentation is subject to change without notice. If you find any problems in this product or documentation, please report them to us in
writing.
Informatica, Informatica Platform, Informatica Data Services, PowerCenter, PowerCenterRT, PowerCenter Connect, PowerCenter Data Analyzer, PowerExchange,
PowerMart, Metadata Manager, Informatica Data Quality, Informatica Data Explorer, Informatica B2B Data Transformation, Informatica B2B Data Exchange Informatica On
Demand, Informatica Identity Resolution, Informatica Application Information Lifecycle Management, Informatica Complex Event Processing, Ultra Messaging and Informatica
Master Data Management are trademarks or registered trademarks of Informatica Corporation in the United States and in jurisdictions throughout the world. All other company
and product names may be trade names or trademarks of their respective owners.
Portions of this software and/or documentation are subject to copyright held by third parties, including without limitation: Copyright DataDirect Technologies. All rights
reserved. Copyright

Sun Microsystems. All rights reserved. Copyright

RSA Security Inc. All Rights Reserved. Copyright

Ordinal Technology Corp. All rights


reserved.Copyright

Aandacht c.v. All rights reserved. Copyright Genivia, Inc. All rights reserved. Copyright Isomorphic Software. All rights reserved. Copyright

Meta
Integration Technology, Inc. All rights reserved. Copyright

Intalio. All rights reserved. Copyright

Oracle. All rights reserved. Copyright

Adobe Systems Incorporated. All


rights reserved. Copyright

DataArt, Inc. All rights reserved. Copyright

ComponentSource. All rights reserved. Copyright

Microsoft Corporation. All rights reserved.


Copyright

Rogue Wave Software, Inc. All rights reserved. Copyright

Teradata Corporation. All rights reserved. Copyright

Yahoo! Inc. All rights reserved. Copyright

Glyph & Cog, LLC. All rights reserved. Copyright

Thinkmap, Inc. All rights reserved. Copyright

Clearpace Software Limited. All rights reserved. Copyright

Information
Builders, Inc. All rights reserved. Copyright

OSS Nokalva, Inc. All rights reserved. Copyright Edifecs, Inc. All rights reserved. Copyright Cleo Communications, Inc. All rights
reserved. Copyright

International Organization for Standardization 1986. All rights reserved. Copyright

ej-technologies GmbH. All rights reserved. Copyright

Jaspersoft
Corporation. All rights reserved. Copyright

is International Business Machines Corporation. All rights reserved. Copyright

yWorks GmbH. All rights reserved. Copyright

Lucent Technologies. All rights reserved. Copyright (c) University of Toronto. All rights reserved. Copyright

Daniel Veillard. All rights reserved. Copyright

Unicode, Inc.
Copyright IBM Corp. All rights reserved. Copyright

MicroQuill Software Publishing, Inc. All rights reserved. Copyright

PassMark Software Pty Ltd. All rights reserved.


Copyright

LogiXML, Inc. All rights reserved. Copyright

2003-2010 Lorenzi Davide, All rights reserved. Copyright

Red Hat, Inc. All rights reserved. Copyright

The Board
of Trustees of the Leland Stanford Junior University. All rights reserved. Copyright

EMC Corporation. All rights reserved. Copyright

Flexera Software. All rights reserved.


This product includes software developed by the Apache Software Foundation (http://www.apache.org/), and other software which is licensed under the Apache License,
Version 2.0 (the "License"). You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0. Unless required by applicable law or agreed to in writing,
software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
License for the specific language governing permissions and limitations under the License.
This product includes software which was developed by Mozilla (http://www.mozilla.org/), software copyright The JBoss Group, LLC, all rights reserved; software copyright

1999-2006 by Bruno Lowagie and Paulo Soares and other software which is licensed under the GNU Lesser General Public License Agreement, which may be found at http://
www.gnu.org/licenses/lgpl.html. The materials are provided free of charge by Informatica, "as-is", without warranty of any kind, either express or implied, including but not
limited to the implied warranties of merchantability and fitness for a particular purpose.
The product includes ACE(TM) and TAO(TM) software copyrighted by Douglas C. Schmidt and his research group at Washington University, University of California, Irvine,
and Vanderbilt University, Copyright (

) 1993-2006, all rights reserved.


This product includes software developed by the OpenSSL Project for use in the OpenSSL Toolkit (copyright The OpenSSL Project. All Rights Reserved) and redistribution of
this software is subject to terms available at http://www.openssl.org and http://www.openssl.org/source/license.html.
This product includes Curl software which is Copyright 1996-2007, Daniel Stenberg, <daniel@haxx.se>. All Rights Reserved. Permissions and limitations regarding this
software are subject to terms available at http://curl.haxx.se/docs/copyright.html. Permission to use, copy, modify, and distribute this software for any purpose with or without
fee is hereby granted, provided that the above copyright notice and this permission notice appear in all copies.
The product includes software copyright 2001-2005 (

) MetaStuff, Ltd. All Rights Reserved. Permissions and limitations regarding this software are subject to terms available
at http://www.dom4j.org/ license.html.
The product includes software copyright

2004-2007, The Dojo Foundation. All Rights Reserved. Permissions and limitations regarding this software are subject to terms
available at http://dojotoolkit.org/license.
This product includes ICU software which is copyright International Business Machines Corporation and others. All rights reserved. Permissions and limitations regarding this
software are subject to terms available at http://source.icu-project.org/repos/icu/icu/trunk/license.html.
This product includes software copyright

1996-2006 Per Bothner. All rights reserved. Your right to use such materials is set forth in the license which may be found at http://
www.gnu.org/software/ kawa/Software-License.html.
This product includes OSSP UUID software which is Copyright

2002 Ralf S. Engelschall, Copyright

2002 The OSSP Project Copyright

2002 Cable & Wireless


Deutschland. Permissions and limitations regarding this software are subject to terms available at http://www.opensource.org/licenses/mit-license.php.
This product includes software developed by Boost (http://www.boost.org/) or under the Boost software license. Permissions and limitations regarding this software are subject
to terms available at http:/ /www.boost.org/LICENSE_1_0.txt.
This product includes software copyright

1997-2007 University of Cambridge. Permissions and limitations regarding this software are subject to terms available at http://
www.pcre.org/license.txt.
This product includes software copyright

2007 The Eclipse Foundation. All Rights Reserved. Permissions and limitations regarding this software are subject to terms
available at http:// www.eclipse.org/org/documents/epl-v10.php.
This product includes software licensed under the terms at http://www.tcl.tk/software/tcltk/license.html, http://www.bosrup.com/web/overlib/?License, http://www.stlport.org/
doc/ license.html, http://www.asm.ow2.org/license.html, http://www.cryptix.org/LICENSE.TXT, http://hsqldb.org/web/hsqlLicense.html, http://httpunit.sourceforge.net/doc/
license.html, http://jung.sourceforge.net/license.txt , http://www.gzip.org/zlib/zlib_license.html, http://www.openldap.org/software/release/license.html, http://www.libssh2.org,
http://slf4j.org/license.html, http://www.sente.ch/software/OpenSourceLicense.html, http://fusesource.com/downloads/license-agreements/fuse-message-broker-v-5-3- license-
agreement; http://antlr.org/license.html; http://aopalliance.sourceforge.net/; http://www.bouncycastle.org/licence.html; http://www.jgraph.com/jgraphdownload.html; http://
www.jcraft.com/jsch/LICENSE.txt. http://jotm.objectweb.org/bsd_license.html; . http://www.w3.org/Consortium/Legal/2002/copyright-software-20021231; http://www.slf4j.org/
license.html; http://developer.apple.com/library/mac/#samplecode/HelpHook/Listings/HelpHook_java.html; http://nanoxml.sourceforge.net/orig/copyright.html; http://
www.json.org/license.html; http://forge.ow2.org/projects/javaservice/, http://www.postgresql.org/about/licence.html, http://www.sqlite.org/copyright.html, http://www.tcl.tk/
software/tcltk/license.html, http://www.jaxen.org/faq.html, http://www.jdom.org/docs/faq.html, http://www.slf4j.org/license.html; http://www.iodbc.org/dataspace/iodbc/wiki/
iODBC/License; http://www.keplerproject.org/md5/license.html; http://www.toedter.com/en/jcalendar/license.html; http://www.edankert.com/bounce/index.html; http://www.net-
snmp.org/about/license.html; http://www.openmdx.org/#FAQ; http://www.php.net/license/3_01.txt; http://srp.stanford.edu/license.txt; http://www.schneier.com/blowfish.html;
http://www.jmock.org/license.html; http://xsom.java.net; and http://benalman.com/about/license/.
This product includes software licensed under the Academic Free License (http://www.opensource.org/licenses/afl-3.0.php), the Common Development and Distribution
License (http://www.opensource.org/licenses/cddl1.php) the Common Public License (http://www.opensource.org/licenses/cpl1.0.php), the Sun Binary Code License
Agreement Supplemental License Terms, the BSD License (http:// www.opensource.org/licenses/bsd-license.php) the MIT License (http://www.opensource.org/licenses/mit-
license.php) and the Artistic License (http://www.opensource.org/licenses/artistic-license-1.0).
This product includes software copyright

2003-2006 Joe WaInes, 2006-2007 XStream Committers. All rights reserved. Permissions and limitations regarding this software
are subject to terms available at http://xstream.codehaus.org/license.html. This product includes software developed by the Indiana University Extreme! Lab. For further
information please visit http://www.extreme.indiana.edu/.
This product includes software developed by Andrew Kachites McCallum. "MALLET: A Machine Learning for Language Toolkit." http://mallet.cs.umass.edu (2002).
This Software is protected by U.S. Patent Numbers 5,794,246; 6,014,670; 6,016,501; 6,029,178; 6,032,158; 6,035,307; 6,044,374; 6,092,086; 6,208,990; 6,339,775;
6,640,226; 6,789,096; 6,820,077; 6,823,373; 6,850,947; 6,895,471; 7,117,215; 7,162,643; 7,243,110, 7,254,590; 7,281,001; 7,421,458; 7,496,588; 7,523,121; 7,584,422;
7676516; 7,720,842; 7,721,270; and 7,774,791, international Patents and other Patents Pending.
DISCLAIMER: Informatica Corporation provides this documentation "as is" without warranty of any kind, either express or implied, including, but not limited to, the implied
warranties of noninfringement, merchantability, or use for a particular purpose. Informatica Corporation does not warrant that this software or documentation is error free. The
information provided in this software or documentation may include technical inaccuracies or typographical errors. The information in this software and documentation is
subject to change at any time without notice.
NOTICES
This Informatica product (the "Software") includes certain drivers (the "DataDirect Drivers") from DataDirect Technologies, an operating company of Progress Software
Corporation ("DataDirect") which are subject to the following terms and conditions:
1. THE DATADIRECT DRIVERS ARE PROVIDED "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT.
2. IN NO EVENT WILL DATADIRECT OR ITS THIRD PARTY SUPPLIERS BE LIABLE TO THE END-USER CUSTOMER FOR ANY DIRECT, INDIRECT,
INCIDENTAL, SPECIAL, CONSEQUENTIAL OR OTHER DAMAGES ARISING OUT OF THE USE OF THE ODBC DRIVERS, WHETHER OR NOT INFORMED OF
THE POSSIBILITIES OF DAMAGES IN ADVANCE. THESE LIMITATIONS APPLY TO ALL CAUSES OF ACTION, INCLUDING, WITHOUT LIMITATION, BREACH
OF CONTRACT, BREACH OF WARRANTY, NEGLIGENCE, STRICT LIABILITY, MISREPRESENTATION AND OTHER TORTS.
Part Number: PC-BDE-95100-0000
Table of Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
Informatica Resources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
Informatica Customer Portal. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
Informatica Documentation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
Informatica Web Site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
Informatica How-To Library. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
Informatica Knowledge Base. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Informatica Multimedia Knowledge Base. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Informatica Global Customer Support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Chapter 1: Introduction to PowerCenter Big Data Edition. . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PowerCenter Big Data Edition Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Big Data Access. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Data Replication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
High-Performance Processing in the Native Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Native Environment Processing Architecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
High-Performance Processing in a Hive Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Hive Environment Processing Architecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Big Data Processing Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Chapter 2: Installation and Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Installation and Configuration Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
PowerCenter Big Data Edition Installation Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Before You Begin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Install and Configure PowerCenter Standard Edition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Install and Configure PowerExchange Adapters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Install and Configure Data Replication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Pre-Installation Tasks for a Single Node Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Pre-Installation Tasks for a Cluster Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
PowerCenter Big Data Edition Installation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Installing in a Single Node Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Installing in a Cluster Environment from the Primary NameNode Using SCP Protocol. . . . . . . . . . 11
Installing in a Cluster Environment from the Primary NameNode Using FTP, HTTP, or NFS
Protocol. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Installing in a Cluster Environment from any Machine. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
After You Install. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Hadoop Environment Variable Properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Hadoop Pushdown Properties for the Data Integration Service. . . . . . . . . . . . . . . . . . . . . . . . . 14
Table of Contents i
HDFS Security Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Set Up Address Validation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
PowerCenter Big Data Edition Uninstallation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Uninstalling PowerCenter Big Data Edition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Chapter 3: Connections. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Connections Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
HDFS Connection Properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Hive Connection Properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Creating a Connection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Chapter 4: Mappings in the Native Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Mappings in the Native Environment Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Data Processor Mappings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
HDFS Mappings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
HDFS Mapping Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Hive Mappings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Hive Mapping Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Social Media Mappings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Twitter Mapping Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Chapter 5: Mappings in a Hive Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Mappings in a Hive Environment Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Datatypes in a Hive Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Sources in a Hive Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Flat File Sources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Hive Sources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Relational Sources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Targets in a Hive Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Flat File Targets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
HDFS Flat File Targets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Hive Targets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Relational Targets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Transformations in a Hive Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Functions in a Hive Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Variable Ports in a Hive Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Mappings in a Hive Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Workflows that Run Mappings in a Hive Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Configuring a Mapping to Run in a Hive Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Hive Execution Plan. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Hive Execution Plan Details. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Viewing the Hive Execution Plan for a Mapping. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Monitoring a Mapping. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
ii Table of Contents
Logs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Troubleshooting a Mapping in a Hive Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Chapter 6: Profiles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Profiles Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Native and Hadoop Environments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Supported Data Source and Run-time Environments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Run-time Environment Setup and Validation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Run-time Environment and Profile Performance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Profile Types on Hadoop. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Column Profiles on Hadoop. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Rule Profiles on Hadoop. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Data Domain Discovery on Hadoop. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Running a Single Data Object Profile on Hadoop. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Running Multiple Data Object Profiles on Hadoop. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Monitoring a Profile. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Viewing Profile Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Troubleshooting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Chapter 7: Native Environment Optimization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Native Environment Optimization Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Processing Big Data on a Grid. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Data Integration Service Grid. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
PowerCenter Integration Service Grid. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Grid Optimization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Processing Big Data on Partitions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Partition Optimization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
High Availability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Appendix A: Datatype Reference. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Datatype Reference Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Hive Complex Datatypes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Hive Datatypes and Transformation Datatypes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Appendix B: Glossary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Table of Contents iii
Preface
The Informatica for Hadoop User Guide provides information about how to configure Informatica products for
Hadoop.
Informatica Resources
Informatica Customer Portal
As an Informatica customer, you can access the Informatica Customer Portal site at
http://mysupport.informatica.com. The site contains product information, user group information, newsletters,
access to the Informatica customer support case management system (ATLAS), the Informatica How-To Library,
the Informatica Knowledge Base, the Informatica Multimedia Knowledge Base, Informatica Product
Documentation, and access to the Informatica user community.
Informatica Documentation
The Informatica Documentation team takes every effort to create accurate, usable documentation. If you have
questions, comments, or ideas about this documentation, contact the Informatica Documentation team through
email at infa_documentation@informatica.com. We will use your feedback to improve our documentation. Let us
know if we can contact you regarding your comments.
The Documentation team updates documentation as needed. To get the latest documentation for your product,
navigate to Product Documentation from http://mysupport.informatica.com.
Informatica Web Site
You can access the Informatica corporate web site at http://www.informatica.com. The site contains information
about Informatica, its background, upcoming events, and sales offices. You will also find product and partner
information. The services area of the site includes important information about technical support, training and
education, and implementation services.
Informatica How-To Library
As an Informatica customer, you can access the Informatica How-To Library at http://mysupport.informatica.com.
The How-To Library is a collection of resources to help you learn more about Informatica products and features. It
includes articles and interactive demonstrations that provide solutions to common problems, compare features and
behaviors, and guide you through performing specific real-world tasks.
iv
Informatica Knowledge Base
As an Informatica customer, you can access the Informatica Knowledge Base at http://mysupport.informatica.com.
Use the Knowledge Base to search for documented solutions to known technical issues about Informatica
products. You can also find answers to frequently asked questions, technical white papers, and technical tips. If
you have questions, comments, or ideas about the Knowledge Base, contact the Informatica Knowledge Base
team through email at KB_Feedback@informatica.com.
Informatica Multimedia Knowledge Base
As an Informatica customer, you can access the Informatica Multimedia Knowledge Base at
http://mysupport.informatica.com. The Multimedia Knowledge Base is a collection of instructional multimedia files
that help you learn about common concepts and guide you through performing specific tasks. If you have
questions, comments, or ideas about the Multimedia Knowledge Base, contact the Informatica Knowledge Base
team through email at KB_Feedback@informatica.com.
Informatica Global Customer Support
You can contact a Customer Support Center by telephone or through the Online Support. Online Support requires
a user name and password. You can request a user name and password at http://mysupport.informatica.com.
Use the following telephone numbers to contact Informatica Global Customer Support:
North America / South America Europe / Middle East / Africa Asia / Australia
Toll Free
Brazil: 0800 891 0202
Mexico: 001 888 209 8853
North America: +1 877 463 2435
Toll Free
France: 0805 804632
Germany: 0800 5891281
Italy: 800 915 985
Netherlands: 0800 2300001
Portugal: 800 208 360
Spain: 900 813 166
Switzerland: 0800 463 200
United Kingdom: 0800 023 4632


Standard Rate
Belgium: +31 30 6022 797
France: +33 1 4138 9226
Germany: +49 1805 702 702
Netherlands: +31 306 022 797
United Kingdom: +44 1628 511445
Toll Free
Australia: 1 800 151 830
New Zealand: 09 9 128 901


Standard Rate
India: +91 80 4112 5738
Preface v
vi
C H A P T E R 1
Introduction to PowerCenter Big
Data Edition
This chapter includes the following topics:
PowerCenter Big Data Edition Overview, 1
Big Data Access, 2
Data Replication, 2
High-Performance Processing in the Native Environment, 3
High-Performance Processing in a Hive Environment, 4
Big Data Processing Example, 5
PowerCenter Big Data Edition Overview
PowerCenter Big Data Edition includes functionality from the following Informatica products: PowerCenter, Data
Explorer, Data Quality, Data Replication, Data Tranformation, PowerExchange for Hive, PowerExchange for
HDFS, PowerExchange for Hadoop, and social media adapters.
In addition to basic functionality associated with the Informatica products, you can use the following functionality
associated with big data:
Access big data sources
Access unstructured and semi-structured data, social media data, and data in Hive and HDFS.
Replicate data
Replicate large amounts of transactional data between heterogeneous databases and platforms.
Configure high-performance processing in the native environment
Distribute mapping, session, and workflow processing across nodes in a grid, enable partitioning to process
partitions of data in parallel, and process data through highly available application services in the domain.
Configure high-performance processing in a Hive environment
Distribute mapping and profile processing across cluster nodes in a Hive environment.
You can process data in the native environment or a Hive environment. In the native environment, an Integration
Service processes the data. You can run Model repository mappings and profiles on the Data Integration Service.
You can run PowerCenter sessions and workflows on a PowerCenter Integration Service. In a Hive environment,
nodes in a Hadoop cluster process the data.
1
Big Data Access
In addition to relational and flat file data, you can access unstructured and semi-structured data, social media
data, and data in a Hive or Hadoop Distributed File System (HDFS) environment.
You can access the following types of data:
Transaction data
You can access different types of transaction data, including data from relational database management
systems, online transaction processing systems, online analytical processing systems, enterprise resource
planning systems, customer relationship managment systems, mainframe, and cloud.
Unstructured and semi-strutured data
You can use parser transformations to read and transform unstructured and semi-structured data. For
example, you can use the Data Processor transformation in a workflow to parse a Microsoft Word file to load
customer and order data into relational database tables.
You can use HParser to transform complex data into flattened, usable formats for Hive, PIG, and MapReduce
processing. HParser processes complex files, such as messaging formats, HTML pages and PDF documents.
HParser also transforms formats such as ACORD, HIPAA, HL7, EDI-X12, EDIFACT, AFP, and SWIFT.
Social media data
You can use PowerExchange adapters for social media to read data from social media web sites like
Facebook, Twitter, and LinkedIn. You can also use the PowerExchange for DataSift to extract real-time data
from different social media web sites and capture data from DataSift regarding sentiment and language
analysis. You can use PowerExchange for Web Content-Kapow to extract data from any web site.
Data in Hive and HDFS
You can use other PowerExchange adapters to read data from or write data to Hadoop. For example, you can
use PowerExchange for Hive to read data from or write data to Hive. Also, you can use PowerExchange for
HDFS to extract data from and load data to HDFS.
Data Replication
You can replicate large amounts of transactional data between heterogeneous databases and platforms with Data
Replication. You might replicate data to distribute or migrate the data across your environment.
With Data Replication, you can perform the following types of data replication:
Low-latency data replication
You can perform low-latency batched replication to replicate data on an interval. You can also perform
continuous replication to replicate data in near real time.
For example, you can use continuous replication to send transactional changes to a staging database or
operational data store. You can then use PowerCenter to extract data from Data Replication target tables and
then transform the data before loading it to an active enterprise data warehouse.
Data replication for Hadoop processing
You can extract transactional changes into text files. You can then use PowerCenter to move the text files to
Hadoop to be processed.
2 Chapter 1: Introduction to PowerCenter Big Data Edition
High-Performance Processing in the Native
Environment
You can optimize the native environment to process big data fast and reliably. You can run an Integration Service
on a grid to distribute the processing across nodes in the grid. You can process partitions of a session in parallel.
You can also enable high availability.
You can enable the following features to optimize the native environment:
PowerCenter Integration Service on grid
You can run PowerCenter sessions and workflows on a grid. The grid is an alias assigned to a group of nodes
that run PowerCenter sessions and workflows. When you run a session or workflow on a grid, the
PowerCenter Integration Service distributes the processing across multiple nodes in the grid.
Data Integration Service on grid
You can run Model repository mappings and profiles on a grid. The grid is an alias assigned to a group of
nodes that run mappings and profiles assigned to the Data Integration Service. When you run a mapping or
profile on a grid, the Data Integration Service distributes the processing across multiple nodes in the grid.
Partitioning
You can create partitions in a PowerCenter session to increase performance. When you run a partitioned
session, the PowerCenter Integration Service performs the extract, transformation, and load for each partition
in parallel.
High availability
You can enable high availability to eliminate single points of failure for PowerCenter application services.
PowerCenter application services can continue running despite temporary network or hardware failures.
For example, if you run the PowerCenter Integration Service on a grid and one of the nodes becomes
unavailable, the PowerCenter Integration Service recovers the tasks and runs them on a different node. If you
run the PowerCenter Integration Service on a single node and you enable high availability, you can configure
backup nodes in case the primary node becomes unavailable.
Native Environment Processing Architecture
You can run sessions, profiles, and workflows on an Integration Service grid. You can run PowerCenter sessions
and workflows on a PowerCenter Integration Service grid. You can run Model repository profiles and workflows on
a Data Integration Service grid.
The following diagram shows the service process distribution when you run a PowerCenter workflow on a
PowerCenter Integration Service grid with three nodes:
High-Performance Processing in the Native Environment 3
When you run the workflow on a grid, the PowerCenter Integration Service process distributes the tasks in the
following way:
On Node 1, the master service process starts the workflow and runs workflow tasks other than the Session,
Command, and predefined Event-Wait tasks. The Load Balancer dispatches the Session, Command, and
predefined Event-Wait tasks to other nodes.
On Node 2, the worker service process starts a process to run a Command task and starts a DTM process to
run Session task 1.
On Node 3, the worker service process runs a predefined Event-Wait task and starts a DTM process to run
Session task 2.
If the master service process becomes unavailable while running a workflow, the PowerCenter Integration Service
can recover the workflow based on the workflow state and recovery strategy. If the workflow was enabled for high
availability recovery, the PowerCenter Integration Service restores the state of operation for the workflow and
recovers the workflow from the point of interruption.
If a worker service process becomes unavailable while running tasks of a workflow, the master service process
can recover tasks based on task state and recovery strategy.
High-Performance Processing in a Hive Environment
You can run Model repository mappings and profiles in a Hive environment to process large amounts of data of 10
terabytes or more. In the Hive environment, the Data Integration Service converts the mapping or profile into
MapReduce programs to enable the Hadoop cluster to process the data.
Hive Environment Processing Architecture
You can run Model repository mappings or profiles in a Hive environment.
To run a mapping or profile in a Hive environment, the Data Integration Service creates HiveQL queries based on
the transformation or profiling logic. The Data Integration Service submits the HiveQL queries to the Hive driver.
The Hive driver converts the HiveQL queries to MapReduce jobs, and then sends the jobs to the Hadoop cluster.
The following diagram shows the architecture of how a Hadoop cluster processes MapReduce jobs sent from the
Hive driver:
4 Chapter 1: Introduction to PowerCenter Big Data Edition
The following events occur when the Hive driver sends MapReduce jobs to the Hadoop cluster:
1. The Hive driver sends the MapReduce jobs to the Job Tracker in the Hive environment.
2. The JobTracker retrieves a list of TaskTracker nodes that can process the MapReduce jobs from the
NameNode.
3. The JobTracker assigns MapReduce jobs to TaskTracker nodes.
4. The Hive driver also connects to the Hive metadata database through the Hive metastore to determine where
to create temporary tables. The Hive driver uses temporary tables to process the data. The Hive driver
removes temporary tables after completing the task.
Big Data Processing Example
Every week, an investment banking organization manually calculates the popularity and risk of stocks, and then
matches stocks to each customer based on the preferences of the customer. However, the organization now
wants you to automate this process.
You use the Developer tool to create a workflow that calculates the popularity and risk of each stock, matches
stocks to each customer, and then sends an email with a list of stock recommendations for all customers. To
determine the popularity of a stock, you count the number of times that the stock was included in Twitter feeds and
the number of times customers inquired about the stock on the company stock trade web site.
The following diagram shows the components of the workflow:
You configure the workflow to complete the following tasks:
Big Data Processing Example 5
1. Extract and count the number of inquiries about stocks from weblogs.
Extracts the inquiries about each stock from the weblogs, and then counts the number of inquiries about each
stock. The weblogs are from the company stock trade web site.
2. Extract and count the number of tweets for each stock from Twitter.
Extracts tweets from Twitter, and then counts the number of tweets about each stock.
3. Extract market data and calculate the risk of each stock based on market data.
Extracts the daily high stock value, daily low stock value, and volatility of each stock from a flat file provided
by a third-party vendor. The workflow calculates the risk of each stock based on the extracted market data.
4. Combine the inquiry count, tweet count, and risk for each stock.
Combines the inquiry count, tweet count, and risk for each stock from the weblogs, Twitter, and market data,
respectively.
5. Extract historical stock transactions for each customer.
Extracts historical stock purchases of each customer from a database.
6. Calculate the average risk and average popularity of the stocks purchased by each customer.
Calculates the average risk and average popularity of all stocks purchased by each customer.
7. Match stocks to each customer based on their preferences.
Matches stocks that have the same popularity and risk as the average popularity and average risk of the
stocks that the customer previously purchased.
8. Load stock recommendations into the data warehouse.
Loads the stock recommendations into data warehouse to retain a history of the recommendations.
9. Send an email with stock recommendations.
Consolidates the stock recommendations for all customers, and sends an email with the list of
recommendations.
After you create the workflow, you configure it to run in a Hive environment because the workflow must process 15
terabytes of data each time it creates recommendations for customers.
6 Chapter 1: Introduction to PowerCenter Big Data Edition
C H A P T E R 2
Installation and Configuration
This chapter includes the following topics:
Installation and Configuration Overview, 7
Before You Begin, 8
PowerCenter Big Data Edition Installation, 11
After You Install, 13
PowerCenter Big Data Edition Uninstallation, 18
Installation and Configuration Overview
The PowerCenter Big Data Edition installation is distributed as a RedHat Package Manager (RPM) installation
package.
The RPM package includes the Informatica 9.5.1 engine and adapter components. The RPM package and the
binary files needed to run the PowerCenter Big Data Edition installation, are compressed into a tar.gz file.
PowerCenter Big Data Edition Installation Process
You can install PowerCenter Big Data Edition in a single node or cluster environment.
Installing in a Single Node Environment
You can install PowerCenter Big Data Edition in a single node environment.
1. Extract the PowerCenter Big Data Edition tar.gz file to the machine.
2. Install PowerCenter Big Data Edition by running the installation shell script in a Linux environment.
Installing in a Cluster Environment
You can install PowerCenter Big Data Edition in a cluster environment.
1. Extract the PowerCenter Big Data Edition tar.gz file to a machine.
2. Distribute the RPM package to all of the nodes within the Hadoop cluster. You can distribute the RPM
package using any of the following protocols: File Transfer Protocol (FTP), Hypertext Transfer Protocol
(HTTP), Network File System (NFS), or Secure Copy (SCP) protocol.
7
3. Install PowerCenter Big Data Edition by running the installation shell script in a Linux environment. You can
install PowerCenter Big Data Edition from the primary NameNode or from any machine using the
HadoopDataNodes file.
Install from the primary NameNode. You can install PowerCenter Big Data Edition using FTP, HTTP, NFS
or SCP protocol. During the installation, the installer shell script picks up all of the DataNodes from the
$HADOOP_HOME/conf/slaves file and copies the PowerCenter Big Data Edition binary files to the /
<PowerCenterBigDataEditionInstallationDirectory>/Informatica directory on each of the DataNodes. You
can perform this step only if you are deploying Hadoop from the primary NameNode.
Install from any machine. Add the IP addresses or machine host names, one for each line, for each of the
nodes in the Hadoop cluster in the HadoopDataNodes file. During the PowerCenter Big Data Edition
installation, the installation shell script picks up all of the nodes from the HadoopDataNodes file and copies
the PowerCenter Big Data Edition binary files to the /<PowerCenterBigDataEditionInstallationDirectory>/
Informatica directory on each of the nodes.
Before You Begin
Before you begin the PowerCenter Big Data Edition installation, install the PowerCenter components and
PowerExchange adapters, and perform the pre-installation tasks.
Install and Configure PowerCenter Standard Edition
Before you install PowerCenter Big Data Edition, install and configure Informatica 9.5.1 PowerCenter Standard
Edition.
The PowerCenter Standard Edition installation consists of a server component and a client component.
Informatica Services
Run the Informatica services installation to configure the PowerCenter domain and create the Informatica services.
Informatica Clients
Run the Informatica client installation to create the PowerCenter Client.
For information, see the Informatica PowerCenter Installation and Configuration Guide.
Install and Configure PowerExchange Adapters
Based on your business needs, install and configure PowerExchange adapters.
Use PowerCenter Big Data Edition with PowerCenter and Informatica adapters for access to sources and targets.
You must install and configure PowerExchange for Hive to run Informatica mappings in a Hive environment. For
information, see the Informatica PowerExchange for Hive User Guide.
PowerCenter Adapters
Use PowerCenter adapters, such as PowerExchange for Hadoop, to define sources and targets in PowerCenter
mappings.
For more information about installing and configuring PowerCenter adapters, see the PowerExchange adapter
documentation.
8 Chapter 2: Installation and Configuration
Informatica Adapters
You can use the following Informatica adapters as part of PowerCenter Big Data Edition:
PowerExchange for DataSift
PowerExchange for Facebook
PowerExchange for HDFS
PowerExchange for Hive
PowerExchange for LinkedIn
PowerExchange for Teradata Parallel Transporter API
PowerExchange for Twitter
PowerExchange for Web Content-Kapow Katalyst
For more information, see the PowerExchange adapter documentation.
Install and Configure Data Replication
Before you install PowerCenter Big Data Edition, install and configure Data Replication.
To migrate data with minimal downtime and perform auditing and operational reporting functions, install and
configure Data Replication. For information, see the Informatica Data Replication User Guide.
Pre-Installation Tasks for a Single Node Environment
Before you begin the PowerCenter Big Data Edition installation in a single node environment, perform the pre-
installation requirements.
Verify that Hadoop is installed with Hadoop File System (HDFS) and MapReduce. Informatica supports the
Cloudera (CDH Version 3 Update 4) and Apache (Hadoop 1.0.3) Hadoop distributions. Refer to http://
hadoop.apache.org/ for more information. The Hadoop installation should include a Hive data warehouse that is
configured to use a MySQL database as the MetaStore. You can configure Hive to use a local or remote
MetaStore server.
Note: Informatica does not support embedded MetaStore server setups.
Install the required third party client software to perform both read and write operations in native mode. For
example, install the Oracle client to connect to the Oracle database.
Verify that the PowerCenter Big Data Edition administrator user can run sudo commands or have user root
privileges.
Verify that the temporary folder on the local node has at least 700 MB of disk space.
Download the following file to the temporary folder: InformaticaHadoop-<InformaticaForHadoopVersion>.tar.gz
Extract the InformaticaHadoop-<InformaticaForHadoopVersion>.tar.gz file to the local node where you want to
run the PowerCenter Big Data Edition installation.
Before You Begin 9
Pre-Installation Tasks for a Cluster Environment
Before you begin the PowerCenter Big Data Edition installation in a cluster environment, perform the pre-
installation requirements.
Verify that Hadoop is installed on every node within the cluster. Install Hadoop on every node within the cluster.
Verify that Hadoop is installed with Hadoop File System (HDFS) and MapReduce. Informatica supports the
Cloudera (CDH Version 3 Update 4) and Apache (Hadoop 1.0.3) Hadoop distributions. Refer to http://
hadoop.apache.org/ for more information. The Hadoop installation should include a Hive data warehouse that is
configured to use a MySQL database as the MetaStore. You can configure Hive to use a local or remote
MetaStore server.
Note: Informatica does not support embedded MetaStore server setups.
Install the required third party client software to perform both read and write operations in native mode. For
example, install the Oracle client to connect to the Oracle database. Install the third party client software on all
of the nodes within the Hadoop cluster. For Informatica, this is required to run MapReduce jobs.
Verify that the PowerCenter Big Data Edition administrator has user can run sudo commands or have user root
privileges.
Verify that the RPM package can be distributed by File Transfer Protocol (FTP), Hypertext Transfer Protocol
(HTTP), Network File System (NFS), or Secure Copy (SCP) protocol to all of the nodes that are to be included
in the cluster.
If you are installing PowerCenter Big Data Edition in a cluster environment that uses the FTP protocol, verify
that the FTP service is running.
If you are installing PowerCenter Big Data Edition in a cluster environment that uses the HTTP protocol, verify
that the web server is running.
If you are installing PowerCenter Big Data Edition in a cluster environment that uses the SCP protocol, verify
that the SCP service is running.
If you are installing PowerCenter Big Data Edition in a cluster environment, set up password-less Secure Shell
(SSH) connection between the machine where you want to run the PowerCenter Big Data Edition installation
and all of the nodes on which PowerCenter Big Data Edition will be installed.
Verify that the temporary folder in each of the nodes on which PowerCenter Big Data Edition will be installed
has at least 700 MB of temporary disk space.
Download the following file to a temporary folder: InformaticaHadoop-<InformaticaForHadoopVersion>.tar.gz
Copy the following package to a shared directory: InformaticaHadoop-<InformaticaForHadoopVersion>.rpm
For example,
- For HTTP protocol: /var/www/html
- For FTP protocol: /var/ftp/pub
- For NFS: <Shared location on the node. The file location must be accessible by all the nodes in the cluster.>
Note: The RPM package must be stored on local disk and not on HDFS.
Extract the InformaticaHadoop-<InformaticaForHadoopVersion>.tar.gz file to the machine from where you want
to distribute the RPM package and run the PowerCenter Big Data Edition installation.
In the config file on the machine where you want to run the PowerCenter Big Data Edition installation, set
DISTRIBUTOR_NODE to one of the following protocols.
- For FTP protocol, set DISTRIBUTOR_NODE=ftp://<Distributor Node IP Address>/pub
- For HTTP protocol, set DISTRIBUTOR_NODE=http://<Distributor Node IP Address>
- For NFS protocol, set DISTRIBUTOR_NODE=<Shared file location on the node. The file location must be
accessible by all the nodes in the cluster.>
10 Chapter 2: Installation and Configuration
PowerCenter Big Data Edition Installation
You can install PowerCenter Big Data Edition in a single node environment. You can also install PowerCenter Big
Data Edition in a cluster environment from the primary NameNode or from any machine.
Install PowerCenter Big Data Edition in a single node enivironment or cluster environment:
Install PowerCenter Big Data Edition in a single node environment.
Install PowerCenter Big Data Edition in a cluster environment from the primary NameNode using SCP protocol.
Install PowerCenter Big Data Edition in a cluster environment from the primary NameNode using FTP, HTTP,
or NFS protocol.
Install PowerCenter Big Data Edition in a cluster environment from any machine.
Install PowerCenter Big Data Edition from a shell command line.
Installing in a Single Node Environment
You can install PowerCenter Big Data Edition in a single node environment.
1. Log in to the machine.
2. Run the following command from the PowerCenter Big Data Edition root directory to start the installation in
console mode:
bash InformaticaHadoopInstall.sh
3. Press y to accept the PowerCenter Big Data Edition terms of agreement.
4. Press Enter.
5. Press 1 to install PowerCenter Big Data Edition in a single node environment.
6. Press Enter.
7. Type the absolute path for the PowerCenter Big Data Edition installation directory and press Enter.
Start the path with a slash. The directory names in the path must not contain spaces or the following special
characters: { } ! @ # $ % ^ & * ( ) : ; | ' ` < > , ? + [ ] \
If you type a directory path that does not exist, the installer creates the entire directory path on each of the
nodes during the installation. Default is /opt.
8. Press Enter.
The installer creates the /<PowerCenterBigDataEditionInstallationDirectory>/Informatica directory and
populates all of the file systems with the contents of the RPM package.
You can view the informatica-hadoop-install.<DateTimeStamp>.log installation log file to get more information
about the tasks performed by the installer.
Installing in a Cluster Environment from the Primary NameNode Using
SCP Protocol
You can install PowerCenter Big Data Edition in a cluster environment from the primary NameNode using SCP
protocol.
1. Log in to the primary NameNode.
2. Run the following command to start the PowerCenter Big Data Edition installation in console mode:
bash InformaticaHadoopInstall.sh
3. Press y to accept the PowerCenter Big Data Edition terms of agreement.
PowerCenter Big Data Edition Installation 11
4. Press Enter.
5. Press 2 to install PowerCenter Big Data Edition in a cluster environment.
6. Press Enter.
7. Type the absolute path for the PowerCenter Big Data Edition installation directory.
Start the path with a slash. The directory names in the path must not contain spaces or the following special
characters: { } ! @ # $ % ^ & * ( ) : ; | ' ` < > , ? + [ ] \
If you type a directory path that does not exist, the installer creates the entire directory path on each of the
nodes during the installation. Default is /opt.
8. Press Enter.
9. Press 1 to install PowerCenter Big Data Edition from the primary NameNode.
10. Press Enter.
11. Type the absolute path for the Hadoop installation directory. Start the path with a slash.
12. Press Enter.
13. Type y.
14. Press Enter.
The installer retrieves a list of DataNodes from the $HADOOP_HOME/conf/slaves file. On each of the DataNodes,
the installer creates the /<PowerCenterBigDataEditionInstallationDirectory>/Informatica directory and
populates all of the file systems with the contents of the RPM package.
You can view the informatica-hadoop-install.<DateTimeStamp>.log installation log file to get more information
about the tasks performed by the installer.
Installing in a Cluster Environment from the Primary NameNode Using
FTP, HTTP, or NFS Protocol
You can install PowerCenter Big Data Edition in a cluster environment from the primary NameNode using FTP,
HTTP, or NFS protocol.
1. Log in to the primary NameNode.
2. Run the following command to start the PowerCenter Big Data Edition installation in console mode:
bash InformaticaHadoopInstall.sh
3. Press y to accept the PowerCenter Big Data Edition terms of agreement.
4. Press Enter.
5. Press 2 to install PowerCenter Big Data Edition in a cluster environment.
6. Press Enter.
7. Type the absolute path for the PowerCenter Big Data Edition installation directory.
Start the path with a slash. The directory names in the path must not contain spaces or the following special
characters: { } ! @ # $ % ^ & * ( ) : ; | ' ` < > , ? + [ ] \
If you type a directory path that does not exist, the installer creates the entire directory path on each of the
nodes during the installation. Default is /opt.
8. Press Enter.
9. Press 1 to install PowerCenter Big Data Edition from the primary NameNode.
10. Press Enter.
11. Type the absolute path for the Hadoop installation directory. Start the path with a slash.
12. Press Enter.
12 Chapter 2: Installation and Configuration
13. Type n.
14. Press Enter.
15. Type y.
16. Press Enter.
The installer retrieves a list of DataNodes from the $HADOOP_HOME/conf/slaves file. On each of the DataNodes,
the installer creates the /<PowerCenterBigDataEditionInstallationDirectory>/Informatica directory and
populates all of the file systems with the contents of the RPM package.
You can view the informatica-hadoop-install.<DateTimeStamp>.log installation log file to get more information
about the tasks performed by the installer.
Installing in a Cluster Environment from any Machine
You can install PowerCenter Big Data Edition in a cluster environment from any machine.
1. Verify that the PowerCenter Big Data Edition administrator has user root privileges on the node that will be
running the PowerCenter Big Data Edition installation.
2. Log in to the machine as the root user.
3. In the HadoopDataNodes file on the node from where you want to launch the PowerCenter Big Data Edition
installation, add the IP addresses or machine host names, one for each line, of the nodes in the Hadoop
cluster on which you want to install PowerCenter Big Data Edition.
4. Run the following command to start the PowerCenter Big Data Edition installation in console mode:
bash InformaticaHadoopInstall.sh
5. Press y to accept the PowerCenter Big Data Edition terms of agreement.
6. Press Enter.
7. Press 2 to install PowerCenter Big Data Edition in a cluster environment.
8. Press Enter.
9. Type the absolute path for the PowerCenter Big Data Edition installation directory and press Enter. Start the
path with a slash. Default is /opt.
10. Press Enter.
11. Press 2 to install PowerCenter Big Data Edition using the HadoopDataNodes file.
12. Press Enter.
The installer creates the /<PowerCenterBigDataEditionInstallationDirectory>/Informatica directory and
populates all of the file systems with the contents of the RPM package on the first node that appears in the
HadoopDataNodes file. The installer repeats the process for each node in the HadoopDataNodes file.
After You Install
After you install PowerCenter Big Data Edition, perform the post-installation tasks to ensure that PowerCenter Big
Data Edition runs properly.
Complete the following tasks:
Configure the PowerCenter Big Data Edition environment variable properties file.
Configure the Data Integration Service pushdown properties for Hadoop.
After You Install 13
Install the Address Validation reference data.
Hadoop Environment Variable Properties
After you install PowerCenter Big Data Edition, configure the hadoopEnv.properties file to meet the PowerCenter
Big Data Edition requirements.
Configure the hadoopEnv.properties file with the Informatica, locale, and library path environment variables you
want to include in the PowerCenter Big Data Edition environment.
1. Go to the following location: <InformaticaInstallationDir>/services/shared/hadoop/conf
2. Find the file named hadoopEnv.properties.
Back up the file before you modify it.
3. Use a text editor to open the file and modify the properties.
4. Save the properties file with the name hadoopEnv.properties.
Hadoop Pushdown Properties for the Data Integration Service
You must configure Hadoop pushdown properties for the Data Integration Service to run mappings or profiles in a
Hive environment.
You can configure Hadoop pushdown properties for the Data Integration Service from the Administrator tool.
The following table describes the Hadoop pushdown properties for the Data Integration Service:
Property Description
Informatica Home Directory on Hadoop The PowerCenter Big Data Edition home directory on every
data node created by the Hadoop RPM install. Type /
<PowerCenterBigDataEditionInstallationDirectory>/
Informatica.
Hadoop Distribution Directory The directory containing a collection of Hive and Hadoop
JARS on the data nodes from the Hive and Hadoop Install
locations. The directory contains the minimum set of JARS
required to process Informatica mappings in a Hadoop
environment. Type /
<PowerCenterBigDataEditionInstallationDirectory>/
Informatica/services/shared/hadoop/cdh3u4.
You can modify the Hadoop distribution directory on the data
nodes and set this path from the Administrator tool. To specify
a different Hadoop distribution directory:
1. Use the JARS from compatible Hive and Hadoop install
locations.
2. Create a Hadoop distribution directory in the following
directory path:/
<PowerCenterBigDataEditionInstallationDirectory>/
Informatica/services/shared/hadoop/
[Hadoop_distribution_name] or another location.
3. Copy the required Hive JARS from the Hive install location
at /usr/lib/hive/lib to the following directory: /
<PowerCenterBigDataEditionInstallationDirectory>/
Informatica/services/shared/hadoop/
[Hadoop_distribution_name].
4. Copy the required Hadoop JARS from the Hadoop install
location at /usr/lib/hadoop/lib to the following directory: /
14 Chapter 2: Installation and Configuration
Property Description
<PowerCenterBigDataEditionInstallationDirectory>/
Informatica/services/shared/hadoop/
[Hadoop_distribution_name].
5. If you are using Cloudera distribution, copy the required
Snappy libraries from /usr/lib/hadoop/lib/native to the
following directory: /
<PowerCenterBigDataEditionInstallationDirectory>/
Informatica/services/shared/hadoop/
[Hadoop_distribution_name]/lib/native.
Data Integration Service Hadoop Distribution Directory The Hadoop distribution directory on the Data Integration
Service node. The contents of the Data Integration Service
Hadoop distribution directory must be identical to Hadoop
distribution directory on the data nodes.
Hadoop Distribution Directory
You can modify the Hadoop distribution directory on the data nodes.
When you modify the Hadoop distribution directory, you must copy the minimum set of Hive and Hadoop JARS,
and the Snappy libraries required to process Informatica mappings in a Hive environment from your Hadoop install
location. The actual Hive and Hadoop JARS can vary depending on the Hadoop distribution version.
After You Install 15
The following table lists the contents of Hadoop distribution directory that are installed with the Hadoop RPM for
Cloudera in /<PowerCenterBigDataEditionInstallationDirectory>/Informatica/services/shared/hadoop/
[Hadoop_distribution_name] directory:
Directory Cloudera Distribution files
conf hive-default.xml
lib The directory must contain the following JARS from the Hive
installation:
- hive-cli-0.7.1-cdh3u4.jar
- hive-exec-0.7.1-cdh3u4.jar
- hive-jdbc-0.7.1-cdh3u4.jar
- hive-metastore-0.7.1-cdh3u4.jar
- hive-serde-0.7.1-cdh3u4.jar
- hive-service-0.7.1-cdh3u4.jar
- hive-shims-0.7.1-cdh3u4.jar
- ant-contrib-1.0b3.jar
- antlr-runtime-3.0.1.jar
- asm-3.1.jar
- commons-cli-1.2.jar
- commons-codec-1.3.jar
- commons-collections-3.2.1.jar
- commons-dbcp-1.4.jar
- commons-lang-2.4.jar
- commons-logging-1.0.4.jar
- commons-logging-api-1.0.4.jar
- commons-pool-1.5.4.jar
- datanucleus-connectionpool-2.0.3.jar
- datanucleus-core-2.0.3.jar
- datanucleus-enhancer-2.0.3.jar
- datanucleus-rdbms-2.0.3.jar
- derby.jar
- jackson-core-asl-1.7.3.jar
- jackson-mapper-asl-1.7.3.jar
- jdo2-api-2.3-ec.jar
- jline-0.9.94.jar
- json.jar
- libthrift.jar
- slf4j-api-1.6.1.jar
- slf4j-log4j12-1.6.1.jar
- thrift-fb303-0.5.0.jar
The directory must contain the following JARS from Hadoop
installation:
- hadoop-core-0.20.2-cdh3u4.jar
- guava-r09-jarjar.jar
lib\native The directory Linux-amd64-64 or Linux-i386-32 must
contain the following libraries:
- libhadoop.a
- libhadoop.la
- libhadoop.so
- libhadoop.so.1
- libhadoop.so.1.0.0
- libsnappy.a
- libsnappy.la
- libsnappy.so
- libsnappy.so.1
- libsnappy.so.1.1.1
16 Chapter 2: Installation and Configuration
Directory Cloudera Distribution files
- libsnappyjava.so
The following table lists the contents of Hadoop distribution directory that are installed with the Hadoop RPM for
Apache in /opt/Informatica/services/shared/hadoop/[Hadoop_distribution_name] directory:
Directory Apache Distribution Files
conf hive-default.xml
lib The directory must contain the following JARS from the Hive
installation:
- hive-cli-0.7.1.jar
- hive-exec-0.7.1.jar
- hive-jdbc-0.7.1.jar
- hive-metastore-0.7.1.jar
- hive-serde-0.7.1.jar
- hive-service-0.7.1.jar
- hive-shims-0.7.1.jar
- ant-contrib-1.0b3.jar
- antlr-runtime-3.0.1.jar
- asm-3.2.jar
- commons-cli-1.2.jar
- commons-codec-1.4.jar
- commons-collections-3.2.1.jar
- commons-dbcp-1.4.jar
- commons-lang-2.4.jar
- commons-logging-1.1.1.jar
- commons-logging-api-1.0.4.jar
- commons-pool-1.5.4.jar
- datanucleus-connectionpool-2.0.3.jar
- datanucleus-core-2.0.3.jar
- datanucleus-enhancer-2.0.3.jar
- datanucleus-rdbms-2.0.3.jar
- derby.jar
- jdo2-api-2.3-ec.jar
- jline-0.9.94.jar
- json.jar
- libthrift.jar
- slf4j-api-1.6.1.jar
- slf4j-log4j12-1.6.1.jar
- thrift-fb303-0.5.0.jar
The directory must contain the following JARS from the
Hadoop installation:
- hadoop-core-1.0.3.jar
- commons-configuration-1.6.jar
- jackson-core-asl-1.8.8.jar
- jackson-mapper-asl-1.8.8.jar
HDFS Security Configuration
You must enable the HDFS security property dfs.permissions in the following location: /usr/lib/hadoop/conf/hdfs-
site.xml. You must create a Hadoop user with the same user name as the Data Integration Service user name in
the cluster nodes using the following commands:
hadoop fs -mkdir /user/
hadoop fs -chown :/user/
After You Install 17
Update the Hive warehouse directory in the following location: $INFA_HOME/services/shared/hadoop/
hadoopEnv.properties
Set Up Address Validation
After you install PowerCenter Big Data Edition, optionally install address reference data files on the DataNodes.
If you use PowerCenter Big Data edition with a Data Quality license, you can push a mapping that validates the
accuracy of postal address records to a Hadoop cluster. The mapping uses address reference data files to validate
the records.
You purchase address reference data files from Informatica on a subscription basis. You can download the current
address reference data files from Informatica at any time during the subscription period.
Installing the Address Reference Data Files
Create an automation script to install the address reference data files on each DataNode in the cluster.
1. Browse to the address reference data files that you downloaded from Informatica.
2. Extract the compressed address reference data files.
3. Stage the files to the NameNode machine or to another machine that can write to the DataNodes.
4. Create an automation script to copy the files to each DataNode.
The default directory for the address reference data files in the Hadoop environment is /reference_data .
If you staged the files on the NameNode, use the slaves file for the Hadoop cluster to identify the
DataNodes.
If you staged the files on another machine, use the Hadoop_Nodes.txt file to identify the DataNodes. You
find this file in the PowerCenter Big Data Edition installation package.
5. Run the script.
The script copies the address reference data files to the DataNodes.
PowerCenter Big Data Edition Uninstallation
The PowerCenter Big Data Edition uninstallation deletes the PowerCenter Big Data Edition binary files from all of
the DataNodes within the Hadoop cluster. Uninstall PowerCenter Big Data Edition from a shell command.
Uninstalling PowerCenter Big Data Edition
To uninstall PowerCenter Big Data Edition in a single node or cluster environment:
1. Verify that the PowerCenter Big Data Edition administrator can run sudo commands.
2. If you are uninstalling PowerCenter Big Data Edition in a cluster environment, set up password-less Secure
Shell (SSH) connection between the machine where you want to run the PowerCenter Big Data Edition
installation and all of the nodes on which PowerCenter Big Data Edition will be uninstalled.
3. If you are uninstalling PowerCenter Big Data Edition in a cluster environment using the HadoopDataNodes file,
verify that the HadoopDataNodes file contains the IP addresses or machine host names, one for each line, of
each of the nodes in the Hadoop cluster from which you want to uninstall PowerCenter Big Data Edition.
18 Chapter 2: Installation and Configuration
4. Log in to the machine. The machine you log into depends on the PowerCenter Big Data Edition environment
and uninstallation method:
If you are uninstalling PowerCenter Big Data Edition in a single node environment, log in to the machine
on which PowerCenter Big Data Edition is installed.
If you are uninstalling PowerCenter Big Data Edition in a cluster environment using the HADOOP_HOME
environment variable, log in to the primary NameNode.
If you are uninstalling PowerCenter Big Data Edition in a cluster environment using the HadoopDataNodes
file, log in to any node.
5. Run the following command to start the PowerCenter Big Data Edition uninstallation in console mode:
bash InformaticaHadoopInstall.sh
6. Press y to accept the PowerCenter Big Data Edition terms of agreement.
7. Press Enter.
8. Select 3 to uninstall PowerCenter Big Data Edition.
9. Press Enter.
10. Select the uninstallation option, depending on the PowerCenter Big Data Edition environment:
Select 1 to uninstall PowerCenter Big Data Edition in a single node environment.
Select 2 to uninstall PowerCenter Big Data Edition in a cluster environment.
11. Press Enter.
12. If you are uninstalling PowerCenter Big Data Edition in a cluster environment, select the uninstallation option,
depending on the uninstallation method:
Select 1 to uninstall PowerCenter Big Data Edition from the primary NameNode.
Select 2 to uninstall PowerCenter Big Data Edition using the HadoopDataNodes file.
13. Press Enter.
14. If you are uninstalling PowerCenter Big Data Edition in a cluster environment from the primary NameNode,
type the absolute path for the Hadoop installation directory. Start the path with a slash.
The uninstaller deletes all of the PowerCenter Big Data Edition binary files from the /
<PowerCenterBigDataEditionInstallationDirectory>/Informatica directory. In a cluster environment, the uninstaller
delete the binary files from all of the nodes within the Hadoop cluster.
PowerCenter Big Data Edition Uninstallation 19
C H A P T E R 3
Connections
This chapter includes the following topics:
Connections Overview, 20
HDFS Connection Properties, 20
Hive Connection Properties, 21
Creating a Connection, 24
Connections Overview
Define the connections you want to use to access data in Hive or HDFS.
You can create the following types of connections:
HDFS connection. Create an HDFS connection to read data from or write data to the Hadoop cluster.
Hive connection. Create a Hive connection to access Hive data or run Informatica mappings in the Hadoop
cluster. Create a Hive connection in the following connection modes:
- Use the Hive connection to access Hive as a source or target. If you want to use Hive as a target, you need to
have the same connection or another Hive connection that is enabled to run mappings in the Hadoop cluster.
You can access Hive as a source if the mapping is enabled for the native or Hive environment. You can
access Hive as a target only if the mapping is run in the Hadoop cluster.
- Use the Hive connection to validate or run an Informatica mapping in the Hadoop cluster. Before you run
mappings in the Hadoop cluster, review the information in this guide about rules and guidelines for mappings
that you can run in the Hadoop cluster.
You can create the connections using the Developer tool, Administrator tool, and infacmd.
Note: For information about creating connections to other sources or targets such as social media web sites or
Teradata, see the respective PowerExchange adapter user guide for information.
HDFS Connection Properties
Use the HDFS connection to access files in the Hadoop Distributed File System.
20
The following table describes the properties for an HDFS connection:
Property Description
Name The name of the connection. The name is not case sensitive
and must be unique within the domain. You can change this
property after you create the connection. It cannot exceed
128 characters, contain spaces, or contain the following
special characters:
~ ` ! $ % ^ & * ( ) - + = { [ } ] | \ : ; " ' < ,
> . ? /
ID String that the Data Integration Service uses to identify the
connection. The ID is not case sensitive. It must be 255
characters or less and must be unique in the domain. You
cannot change this property after you create the connection.
Default value is the connection name.
Description The description of the connection. The description cannot
exceed 765 characters.
Location The domain where you want to create the connection.
Type The connection type. Default is Hadoop File System.
User Name User name to access HDFS.
NameNode URI The URI to access HDFS. The URI must be in the following
format: hdfs://<namenode>:<port>
Where
- <namenode> is the host name or IP address of the
NameNode.
- <port> is the port that the NameNode listens for remote
procedure calls (RPC).
Hive Connection Properties
Use a Hive connection to access data in Hive or to run a mapping in a Hadoop cluster.
General Properties
The following table describes the general properties that you configure for a Hive connection:
Property Description
Name The name of the connection. The name is not case sensitive and must be unique within
the domain. You can change this property after you create the connection. The name
cannot exceed 128 characters, contain spaces, or contain the following special
characters:
~ ` ! $ % ^ & * ( ) - + = { [ } ] | \ : ; " ' < , > . ? /
ID String that the Data Integration Service uses to identify the connection. The ID is not
case sensitive. It must be 255 characters or less and must be unique in the domain. You
Hive Connection Properties 21
Property Description
cannot change this property after you create the connection. Default value is the
connection name.
Description The description of the connection. The description cannot exceed 4000 characters.
Location The domain where you want to create the connection.
Type The connection type. Select Hive.
Connection Modes Hive connection mode. Select at least one of the following options:
- Access Hive as a source or target. Select this option if you want to use the connection
to access the Hive data warehouse. Note that if you want to use Hive as a target, you
need to enable the same connection or another Hive connection to run mappings in
the Hadoop cluster.
- Use Hive to run mappings in Hadoop cluster. Select this option if you want to use the
connection to run mappings in the Hadoop cluster.
You can select both the options. Default is Access Hive as a source or target.
Environment SQL SQL commands to set the Hadoop environment. In a native environment, the Data
Integration Service executes the environment SQL each time it creates a connection to
Hive metastore. If you use a Hive connection to run mappings in a Hadoop cluster, the
Data Integration Service executes the environment SQL at the start of each Hive session.
The following rules and guidelines apply to the usage of environment SQL in both the
connection modes:
- Use the environment SQL to specify Hive queries.
- Use the environment SQL to set the classpath for Hive user-defined functions and
then use either environment SQL or PreSQL to specify the Hive user-defined
functions. You cannot use PreSQL in the data object properties to specify the
classpath. The path must be the fully qualified path to the JAR files used for user-
defined functions. Set the parameter hive.aux.jars.path with all the entries in
infapdo.aux.jars.path and the path to the JAR files for user-defined functions.
- You can also use environment SQL to define Hadoop or Hive parameters that you
intend to use in the PreSQL commands or in custom queries.
If the Hive connection is used to run mappings in the Hadoop cluster, only the
environment SQL of the Hive connection is executed. The different environment SQL
commands for the connections of the Hive source or target are not executed, even if the
Hive sources and targets are on different clusters.
Properties to Access Hive as Source or Target
The following table describes the connection properties that you configure to access Hive as a source or target:
Property Description
Metadata Connection String The JDBC connection URI used to access the metadata from the Hadoop server.
The connection string must be in the following format:
jdbc:hive://<hostname>:<port>/<db>
Where
- hostname is name or IP address of the machine on which the Hive server is running.
- port is the port on which the Hive server is listening.
- db is the database name to which you want to connect. If you do not provide the
database name, the Data Integration Service uses the default database details.
Bypass Hive JDBC Server JDBC driver mode. Select the checkbox to use the embedded mode or embedded JDBC
driver.
22 Chapter 3: Connections
Property Description
To use the JDBC embedded mode, perform the following tasks:
- Verify that Hive client and Informatica Services are installed on the same machine.
- Configure the Hive connection properties to run mappings in the Hadoop cluster.
If you choose the non-embedded mode, you must configure the Data Access Connection
String.
The JDBC embedded mode is preferred to the non-embedded mode.
Data Access Connection String The connection string used to access data from the Hadoop data store. The non-
embedded JDBC mode connection string must be in the following format:
jdbc:hive://<hostname>:<port>/<db>
Where
- hostname is name or IP address of the machine on which the Hive server is running.
- port is the port on which the Hive server is listening. Default is 10000.
- db is the database to which you want to connect. If you do not provide the database
name, the Data Integration Service uses the default database details.
Properties to Run Mappings in the Hadoop Cluster
The following table describes the Hive connection properties that you configure when you want to use the Hive
connection to run Informatica mappings in a Hive environment:
Property Description
Database Name Namespace for tables. Use the name default for tables that do not have a specified
database name.
Default FS URI The URI to access the default Hadoop Distributed File System (HDFS).
The FS URI must be in the following format:
hdfs://<node name>:<port>
Where
- node name is the host name or IP address of the NameNode.
- port is the port on which the NameNode listens for remote procedure calls (RPC).
JobTracker URI The service within Hadoop that submits the MapReduce tasks to specific nodes in the
cluster.
JobTracker URI must be in the following format:
<jobtrackername>:<port>
Where
- jobtrackername is the host name or IP address of the JobTracker.
- port is the port on which the JobTracker listens for remote procedure calls (RPC).
Hive Warehouse Directory on HDFS The absolute HDFS file path of the default database for the warehouse, which is local to
the cluster. For example, the following file path specifies a local warehouse:
/user/hive/warehouse
Metastore Execution Mode Controls whether to connect to a remote metastore or a local metastore. By default, local
is selected. For a local metastore, you must specify the Metastore Database URI, Driver,
Username, and Password. For a remote metastore, you must specify only the Remote
Metastore URI.
Metastore Database URI The JDBC connection URI used to access the data store in a local metastore setup. The
URI must be in the following format:
jdbc:<datastore type>://<node name>:<port>/<database name>
Hive Connection Properties 23
Property Description
where
- node name is the host name or IP address of the data store.
- data store type is the type of the data store.
- port is the port on which the data store listens for remote procedure calls (RPC).
- database name is the name of the database.
For example, the following URI specifies a local metastore that uses MySQL as a data
store:
jdbc:mysql://hostname23:3306/metastore
Metastore Database Driver Driver class name for the JDBC data store. For example, the following class name
specifies a MySQL driver:
com.mysql.jdbc.Driver
Metastore Database Username The metastore database user name.
Metastore Database Password The password for the metastore user name.
Remote Metastore URI The metastore URI used to access metadata in a remote metastore setup. For a remote
metastore, you must specify the Thrift server details.
The URI must be in the following format:
thrift://<hostname>:<port>
Where
- hostname is name or IP address of the Thrift metastore server.
- port is the port on which the Thrift server is listening.
Creating a Connection
Create a connection before you import data objects, preview data, profile data, and run mappings.
1. Click Window > Preferences.
2. Select Informatica > Connections.
3. Expand the domain in the Available Connections list.
4. Select the type of connection that you want to create:
To select a Hive connection, select Database > Hive.
To select an HDFS connection, select File Systems > Hadoop File System.
5. Click Add.
6. Enter a connection name and optional description.
7. Click Next.
24 Chapter 3: Connections
8. Configure the connection properties. For a Hive connection, you must choose the Hive connection mode and
specify the commands for environment SQL. The SQL commands appy to both the connection modes. Select
at least one of the following connection modes:
Option Description
Access Hive as a source or target Use the connection to access Hive data. If you select this option and click Next, the
Properties to Access Hive as a source or target page appears. Configure the
connection strings.
Run mappings in a Hadoop
cluster.
Use the Hive connection to validate and run Informatica mappings in the Hadoop
cluster. If you select this option and click Next, the Properties used to Run
Mappings in the Hadoop Cluster page appears. Configure the properties.
9. Click Test Connection to verify the connection.
You can test a Hive connection that is configured to access Hive data. You cannot test a Hive connection that
is configured to run Informatica mappings in the Hadoop cluster.
10. Click Finish.
Creating a Connection 25
C H A P T E R 4
Mappings in the Native Environment
This chapter includes the following topics:
Mappings in the Native Environment Overview, 26
Data Processor Mappings, 26
HDFS Mappings, 27
Hive Mappings, 28
Social Media Mappings, 29
Mappings in the Native Environment Overview
You can run a mapping in the native or Hive environment. In the native environment, the Data Integration Service
runs the mapping from the Developer tool. You can run standalone mappings or mappings that are a part of a
workflow.
In the native environment, you can read and process data from large unstructured and semi-structured files, Hive,
or social media web sites. You can include the following objects in the mappings:
Hive sources
Flat file sources or targets in the local system or in HDFS
Complex file sources in the local system or in HDFS
Data Processor transformations to process unstructured and semi-structured file formats
Social media sources
You can also import PowerCenter mappings in the Developer tool and run them in the native environment.
Data Processor Mappings
The Data Processor transformation processes unstructured and semi-structured file formats in a mapping. It
converts source data to flat CSV records that MapReduce applications can process.
You can configure the Data Processor transformation to process messaging formats, HTML pages, XML, and PDF
documents. You can also configure it to transform structured formats such as ACORD, HIPAA, HL7, EDI-X12,
EDIFACT, AFP, and SWIFT.
26
For example, an application produces hundreds of data files per second and writes the files to a directory. You can
create a mapping that extracts the files from the directory, passes them to a Data Processor transformation, and
writes the data to a target.
HDFS Mappings
Create an HDFS mapping to read or write to HDFS.
You can read and write fixed-width and delimited file formats. You can read or write compressed files. You can
read text files and binary file formats such as sequence file from HDFS. You can specify the compression format
of the files. You can use the binary stream output of the complex file data object as input to a Data Processor
transformation to parse the file.
You can define the following objects in an HDFS mapping:
Flat file data object or complex file data object operation as the source to read data from HDFS.
Transformations.
Flat file data object as the target to write data to HDFS or any target.
Validate and run the mapping. You can deploy the mapping and run it or add the mapping to a Mapping task in a
workflow.
HDFS Mapping Example
Your organization, HypoMarket Corporation, needs to analyze purchase order details such as customer ID, item
codes, and item quantity. The purchase order details are stored in a semi-structured compressed XML file in
HDFS. The hierarchical data includes a purchase order parent hierarchy level and a customer contact details child
hierarchy level. Create a mapping that reads all the purchase records from the file in HDFS. The mapping must
convert the hierarchical data to relational data and load it in a relational target.
You can use the extracted data for business analytics.
The following figure shows the example mapping:
You can use the following objects in an HDFS mapping:
HDFS input
The input, Read_Complex_File, is a compressed XML file stored in HDFS.
Data Processor Transformation
The Data Processor transformation, Data_Processor_XML_to_Relational, parses the XML file and provides a
relational output.
Relational output
The output, Write_Relational_Data_Object, is a table in an Oracle database.
HDFS Mappings 27
When you run the mapping, the Data Integration Service reads the file in a binary stream and passes it to the Data
Processor transformation. The Data Processor transformation parses the specified file and provides a relational
output. The output is loaded into the relational target.
You can configure the mapping to run in the native or Hive environment.
Complete the following tasks to configure the mapping:
1. Create an HDFS connection to read files from the Hadoop cluster.
2. Create a complex file data object operation. Specify the following parameters:
The file as the resource in the data object.
The file compression format.
The HDFS file location.
3. Optionally, you can specify the input format that the Mapper uses to read the file.
4. Drag and drop the data object operation into a mapping.
5. Create a Data Processor transformation. Configure the following properties in the Data Processor
transformation:
An input port set to buffer input and binary datatype.
Relational output ports depending on the number of columns you want in the relational output. Specify the
port size for the ports. Use an XML schema reference that describes XML hierarchy. Specify the
normalized output you want. For example, you can specify PurchaseOrderNumber_Key as a generated
key that relates the Purchase Orders output group to a Customer Details group.
Create a Streamer object and specify Streamer as a startup component.
6. Create a relational connection to an Oracle database.
7. Import a relational data object.
8. Create a write transformation for the relational data object and add it to the mapping.
Hive Mappings
Based on the mapping environment, you can read data from or write data to Hive.
In a native environment, you can read data from Hive. To read data from Hive, complete the following steps:
1. Create a Hive connection.
2. Configure the Hive connection mode to access Hive as a source or target.
3. Use the Hive connection to create a data object to read from Hive.
4. Add the data object to a mapping and configure the mapping to run in the native environment.
You can write to Hive in a Hive environment. To read data from Hive, complete the following steps:
1. Create a Hive connection.
2. Configure the Hive connection mode to access Hive as a source or target.
3. Use the Hive connection to create a data object to write to Hive.
4. Add the data object to a mapping and configure the mapping to run in the Hive environment.
You can define the following types of objects in a Hive mapping:
A read data object to read data from Hive
28 Chapter 4: Mappings in the Native Environment
Transformations
A target or an SQL data service. You can write to Hive if you run the mapping in a Hadoop cluster.
Validate and run the mapping. You can deploy the mapping and run it or add the mapping to a Mapping task in a
workflow.
Hive Mapping Example
Your organization, HypoMarket Corporation, needs to analyze customer data. Create a mapping that reads all the
customer records. Create an SQL data service to make a virtual database available for end users to query.
You can use the following objects in a Hive mapping:
Hive input
The input file is a Hive table that contains the customer names and contact details.
Create a relational data object. Configure the Hive connection and specify the table that contains the
customer data as a resource for the data object. Drag the data object into a mapping as a read data object.
SQL Data Service output
Create an SQL data service in the Developer tool. To make it available to end users, include it in an
application, and deploy the application to a Data Integration Service. When the application is running, connect
to the SQL data service from a third-party client tool by supplying a connect string.
You can run SQL queries through the client tool to access the customer data.
Social Media Mappings
Create mappings to read social media data from sources such as Facebook and LinkedIn.
You can extract social media data and load them to a target in the native environment only. You can choose to
parse this data or use the data for data mining and analysis.
To process or analyze the data in Hadoop, you must first move the data to a relational or flat file target and then
run the mapping in the Hadoop cluster.
You can use the following Informatica adapters in the Developer tool:
PowerExchange for DataSift
PowerExchange for Facebook
PowerExchange for LinkedIn
PowerExchange for Twitter
PowerExchange for Web Content-Kapow Katalyst
Review the respective PowerExchange adapter documentation for more information.
Twitter Mapping Example
Your organization, Hypomarket Corporation, needs to review all the tweets that mention your product
"HypoBasket" with a positive attitude since the time you released the product in February 2012.
Create a mapping that identifies tweets that contain the word HypoBasket and writes those records to a table.
Social Media Mappings 29
The following figure shows the example mapping:
You can use the following objects in a Twitter mapping:
Twitter input
The mapping source is a Twitter data object that contains the resource Search.
Create a physical data object and add the data object to the mapping. Add the Search resource to the
physical data object. Modify the query parameter with the following query:
QUERY=HypoBasket:)&since:2012-02-01
Sorter transformation
Optionally, sort the data based on the timestamp.
Add a Sorter transformation to the mapping. Specify the timestamp as the sort key with direction as ascending.
Mapping output
Add a relational data object to the mapping as a target.
After you run the mapping, Data Integration Service writes the extracted tweets to the target table. You can use
text analytics and sentiment analysis tools to analyze the tweets.
30 Chapter 4: Mappings in the Native Environment
C H A P T E R 5
Mappings in a Hive Environment
This chapter includes the following topics:
Mappings in a Hive Environment Overview, 31
Datatypes in a Hive Environment, 32
Sources in a Hive Environment, 32
Targets in a Hive Environment, 33
Transformations in a Hive Environment, 35
Functions in a Hive Environment, 38
Variable Ports in a Hive Environment, 39
Mappings in a Hive Environment, 39
Workflows that Run Mappings in a Hive Environment, 40
Configuring a Mapping to Run in a Hive Environment, 40
Hive Execution Plan, 40
Monitoring a Mapping, 41
Logs, 41
Troubleshooting a Mapping in a Hive Environment, 42
Mappings in a Hive Environment Overview
You can run a mapping on a Hadoop cluster. The Data Integration Service can push mappings that are imported
from PowerCenter or developed in the Developer tool to a Hadoop cluster. You can run standalone mappings or
mappings that are a part of a workflow.
When you run a mapping on a Hadoop cluster, you must configure a Hive validation environment, a Hive run-time
environment, and a Hive connection for the mapping. Validate the mapping to ensure you can push the mapping
logic to Hadoop. After you validate a mapping for the Hive environment, you can run the mapping.
To run a mapping on a Hadoop cluster, complete the following steps:
1. In the Developer tool, create a Hive connection.
2. Create a mapping in the Developer tool or import a mapping from PowerCenter.
3. Configure the mapping to run in a Hive environment.
4. Validate the mapping.
5. Optionally, include the mapping in a workflow.
31
6. Run the mapping or workflow.
When you run the mapping, the Data Integration Service converts the mapping to a Hive execution plan that runs
on a Hadoop cluster. You can view the Hive execution plan using the Developer tool or the Administrator tool.
The Data Integration Service has a Hive executor that can process the mapping. The Hive executor simplifies the
mapping to an equivalent mapping with a reduced set of instructions and generates a Hive execution plan. The
Hive execution plan is a series of Hive queries.The Hive execution plan contains tasks to start the mapping, run
the mapping, and clean up the temporary tables and files. You can view the Hive execution plan that the Data
Integration Service generates before you run the mapping.
You can monitor Hive queries and the Hadoop jobs associated with a query in the Administrator tool. The Data
Integration Service logs messages from the DTM, Hive session, and Hive tasks in the runtime log files.
Datatypes in a Hive Environment
Due to the differences between the native environment and a Hive environment, some variations apply in the
processing and validity of datatypes when you push datatypes to a Hive environment.
The following variations apply in datatype processing and validity:
A Binary datatype in a field or an expression function is not valid. If a transformation has a port with a Binary
datatype that is not used in the mapping, you can validate and run the mapping in a Hive environment.
A high precision Decimal datatype is not valid. A mapping is run in low precision mode in a Hive environment.
The results of arithmetic operations on floating point types, such as a Double or a Decimal, can vary up to 0.1
percent between the native environment and a Hive environment.
Hive complex datatypes in a Hive source or Hive target are not valid.
When the Data Integration Service converts a decimal with a precision of 10 and a scale of 3 to a string
datatype and writes to a flat file target, the results can differ between the native environment and a Hive
environment. For example, in a Hive environment, HDFS writes the output string for the decimal 19711025 with
a precision of 10 and a scale of 3 as 1971. In the native environment, the flat file writer sends the output string
for the decimal 19711025 with a precision of 10 and a scale of 3 as 1971.000.
Sources in a Hive Environment
Due to the differences between the native environment and a Hive environment, you can only push certain
sources to a Hive environment. Some of the sources that are valid in mappings in a Hive environment have
restrictions.
You can run mappings with the following sources in a Hive environment:
IBM DB2
Flat file
HDFS complex file
HDFS flat file
Hive
ODBC
32 Chapter 5: Mappings in a Hive Environment
Oracle
Flat File Sources
Flat file sources are valid in mappings in a Hive environment with some restrictions. A mapping with a flat file
source can fail to run in certain cases.
Flat file sources are valid in mappings in a Hive environment with the following restrictions:
You cannot use a command to generate or transform flat file data and send the output to the flat file reader at
runtime.
You cannot use an indirect source type.
The row size in a flat file source cannot exceed 190 MB.
Hive Sources
Hive sources are valid in mappings in a Hive environment with some restrictions.
Hive sources are valid in mappings in a Hive environment with the following restrictions:
The Data Integration Service can run pre-mapping SQL commands against the source database before it reads
from a Hive source. When you run a mapping with a Hive source in a Hive environment, references to local
path in pre-mapping SQL commands are relative to the Data Integration Service node. When you run a
mapping with a Hive source in the native environment, references to local path in pre-mapping SQL commands
are relative to the Hive server node.
A mapping fails to validate when you configure post-mapping SQL commands.The Data Integration Service
does not run post-mapping SQL commands against a Hive source.
A mapping fails to run when you have Unicode characters in a Hive source definition.
Relational Sources
The Data Integration Service does not run pre-mapping SQL commands or post-mapping SQL commands against
relational sources. You cannot validate and run a mapping with PreSQL or PostSQL properties for a relational
source in a Hive environment.
Targets in a Hive Environment
Due to the differences between the native environment and a Hive environment, you can push only certain targets
to a Hive environment. Some of the targets that are valid in mappings in a Hive environment have restrictions.
You can run mappings with the following targets in a Hive environment:
IBM DB2
Flat file
HDFS flat file
Hive
ODBC
Oracle
Teradata
Targets in a Hive Environment 33
Flat File Targets
Flat file targets are valid in mappings in a Hive environment with some restrictions.
Flat file targets are valid in mappings in a Hive environment with the following restrictions:
The Data Integration Service truncates the target files and reject files before writing the data. When you use a
flat file target, you cannot append output data to target files and reject files.
The Data Integration Service can write to a file output for a flat file target. When you have a flat file target in a
mapping, you cannot write data to a command.
HDFS Flat File Targets
HDFS flat file targets are valid in mappings in a Hive environment with some restrictions.
When you use a HDFS flat file target in a mapping, you must specify the full path that includes the output file
directory and file name. The Data Integration Service may generate multiple output files in the output directory
when you run the mapping in a Hive environment.
Hive Targets
Hive targets are valid in mappings in a Hive environment with some restrictions.
Hive targets are valid in mappings in a Hive environment with the following restrictions:
The Data Integration Service does not run pre-mapping or post-mapping SQL commands against the target
database for a Hive target. You cannot validate and run a mapping with PreSQL or PostSQL properties for a
Hive target,
A mapping fails to run if the Hive target definition differs in the number and order of the columns from the
relational table in the Hive database.
The Data Integration Service uses the truncate table to overwrite data to a Hive target. The Data Integration
Service ignores write properties, update override, delete, insert, and update strategy when it writes data to a
Hive target.
A mapping fails to run when you use Unicode characters in a Hive target definition.
Relational Targets
The Data Integration Service does not run pre-mapping SQL commands or post-mapping SQL commands against
relational targets in a Hive environment. You cannot validate and run a mapping with PreSQL or PostSQL
properties for a relational target in a Hive environment.
34 Chapter 5: Mappings in a Hive Environment
Transformations in a Hive Environment
Due to the differences between native and Hive environment only certain transformations are valid or valid with
restrictions in the Hive environment. The Data Integration Service does not process transformations that contain
functions, expressions, datatypes, and variable fields that are not valid in a Hive environment.
The following table describes the rules and guidelines for transformations:
Transformation Rules and Guidelines
Address Validator You can push mapping logic that includes an Address
Validator transformation to Hadoop if you use a Data Quality
product license.
The following limitation applies to Address Validator
transformations:
- An Address Validator transformation does not generate a
certification report when it runs in a mapping on Hadoop. If
you select a certification report option on the
transformation, the mapping validation fails when you
attempt to push transformation logic to Hadoop.
Aggregator An Aggregator transformation with pass-through fields is valid
if they are group-by fields.
Case Converter The Data Integration Service can push a Case Converter
transformation to Hadoop.
Comparison You can push mapping logic that includes a Comparison
transformation to Hadoop if you use a Data Quality product
license.
Consolidation You can push mapping logic that includes a Consolidation
transformation to Hadoop if you use a Data Quality product
license.
The following limitation applies to Consolidation
transformations:
- A Consolidation transformation may process records in a
different order in native and Hadoop environments. The
transformation may identify a different record as the
survivor record in each environment.
Data Processor The following limitations apply when a Data Processor
transformation directly connects to a complex file reader:
- Ports cannot be defined as file.
- Input port must be defined as binary.
- Output port cannot be defined as binary.
- A Streamer must be defined as startup component.
- Pass-through ports cannot be used.
- Additional input ports cannot be used.
The following limitations apply when a mapping has a Data
Processor transformation:
- Ports cannot be defined as file.
- Ports cannot be defined as binary
- Streamer cannot be defined as startup component.
Decision You can push mapping logic that includes a Decision
transformation to Hadoop if you use a Data Quality product
license.
Transformations in a Hive Environment 35
Transformation Rules and Guidelines
Expression An Expression transformation with a user-defined function
returns a null value for rows that have an exception error in
the function.
The Data Integration Service returns an infinite or a NaN (not
a number) value when you push transformation logic to
Hadoop for expressions that result in numerical errors. For
example:
- Divide by zero
- SQRT (negative number)
- ASIN (out-of-bounds number)
In the native environment, the expressions that result in
numerical errors return null values and the rows do not
appear in the output.
Filter The Data Integration Service can push a Filter transformation
to Hadoop.
Java You must copy external JAR files that a Java transformation
requires to the Informatica installation directory in the Hadoop
cluster nodes at the following location:
[$HADOOP_NODE_INFA_HOME]/services/shared/jars/
platform/dtm/
The following limitations apply to transformation scope
property:
- If the transformation scope is set to Transaction, you
cannot validate the Java transformation. The Data
Integration Service cannot apply transformation logic to all
rows in a transaction.
- If transformation scope is set to Row, a Java
transformation is run by mapper script.
- If you select a port for Java partition key, the
transformation scope is set to All Input.
- If transformation scope is set to All Input, a Java
transformation is run by the reducer script and you must
set at least one input field as a group-by field for the
reducer key.
The Java code in the transformation cannot write output to
standard output when you push transformation logic to
Hadoop. The Java code can write output to standard error
which appears in the log files.
Joiner A Joiner transformation cannot contain inequality joins in the
outer join condition.
Key Generator You can push mapping logic that includes a Key Generator
transformation to Hadoop if you use a Data Quality product
license.
Labeler You can push mapping logic that includes a Labeler
transformation to Hadoop when you configure the
transformation to use probabilistic matching techniques.
You can push mapping logic that includes all types of Labeler
configuration if you use a Data Quality product license.
36 Chapter 5: Mappings in a Hive Environment
Transformation Rules and Guidelines
Lookup The following limitations apply to Lookup transformations:
- An unconnected Lookup transformation is not valid.
- You cannot configure an uncached lookup source.
- You cannot configure a persistent lookup cache for the
lookup source.
- You cannot use a Hive source for a relational lookup
source.
- When you run mappings that contain Lookup
transformations, the Data Integration Service creates
lookup cache Jar files. Hive copies the lookup cache JAR
files to the following temporary directory:/tmp/
<user_name>/hive_resources . The Hive parameter
hive.downloaded.resources.dir determines the location
of the temporary directory. You can delete the lookup
cache JAR files specified in the LDTM log after the
mapping completes to retrieve disk space.
Match You can push mapping logic that includes a Match
transformation to Hadoop if you use a Data Quality product
license.
The following limitation applies to Match transformations:
- A Match transformation generates cluster ID values
differently in native and Hadoop environments. In a Hadoop
environment, the transformation appends a group ID value
to the cluster ID.
Merge The Data Integration Service can push a Merge
transformation to Hadoop.
Parser You can push mapping logic that includes a Parser
transformation to Hadoop when you configure the
transformation to use probabilistic matching techniques.
You can push mapping logic that includes all types of Parser
configuration if you use a Data Quality product license.
Rank A comparison is valid if it is case sensitive.
Router The Data Integration Service can push a Router
transformation to Hadoop.
Sorter The Data Integration service ignores the Sorter transformation
when you push mapping logic to Hadoop.
SQL The Data Integration Service can push SQL transformation
logic to Hadoop.
You cannot use a Hive connection.
Standardizer You can push mapping logic that includes a Standardizer
transformation to Hadoop if you use a Data Quality product
license.
Union The custom source code in the transformation cannot write
output to standard output when you push transformation logic
Transformations in a Hive Environment 37
Transformation Rules and Guidelines
to Hadoop. The custom source code can write output to
standard error, that appears in the runtime log files.
Weighted Average You can push mapping logic that includes a Weighted
Average transformation to Hadoop if you use a Data Quality
product license.
Functions in a Hive Environment
Some transformation language functions that are valid in the native environment are not valid or have limitations in
a Hive environment.
The following table describes the functions that are not valid or have limitations in a Hive environment:
Name Limitation
ABORT String argument is not valid.
AES_DECRYPT Not valid
AES_ENCRYPT Not valid
COMPRESS Not valid
CRC32 Not valid
CUME Not valid
DECODE Not valid
DEC_BASE64 Not valid
DECOMPRESS Not valid
ENC_BASE64 Not valid
ERROR String argument is not valid.
FIRST Not valid
LAST Not valid
MAX (Dates) Not valid
MD5 Not valid
MIN (Dates) Not valid
38 Chapter 5: Mappings in a Hive Environment
Name Limitation
MOVINGAVG Not valid
MOVINGSUM Not valid
Variable Ports in a Hive Environment
A transformation that contains a stateful variable port is not valid in a Hive environment.
A stateful variable port refers to values from previous rows.
Mappings in a Hive Environment
You can run mappings in a Hive environment. Some differences in processing and configuration apply when you
run mappings in a Hive environment.
The following processing differences apply to mappings in a Hive environment:
A mapping is run in low precision mode. The Data Integration Service ignores high precision mode in a Hive
environment. Mappings that require high precision mode may fail to run in a Hive environment.
In a Hive environment, sources that have data errors in a column result in a null value for the column. In the
native environment, the Data Integration Service does not process the rows that have data errors in a column.
When you cancel a mapping that reads from a flat file source, the file copy process that copies flat file data to
HDFS may continue to run. The Data Integration Service logs the command to kill this process in the Hive
session log, and cleans up any data copied to HDFS. Optionally, you can run the command to kill the file copy
process.
The following configuration differences apply to mappings in a Hive environment:
Set the optimizer level to none or minimal if a mapping validates but fails to run. If you set the optimizer level to
use cost-based or semi-join optimization methods, the Data Integration Service ignores this at run-time and
uses the default.
Mappings that contain a Hive source or a Hive target must use the same Hive connection to push the mapping
to Hadoop.
The Data Integration Service ignores the data file block size configured for HDFS files in the hdfs-site.xml file.
The Data Integration Service uses a default data file block size of 64 MB for HDFS files. To change the data file
block size, copy /usr/lib/hadoop/conf/hdfs-site.xml to the following location in the Hadoop distribution
directory for the Data Integration Service node: /opt/Informatica/services/shared/hadoop/
[Hadoop_distribution_name]/conf. You can also update the data file block size in the following file: /opt/
Informatica/services/shared/hadoop/[Hadoop_distribution_name]/conf/hive-default.xml.
Variable Ports in a Hive Environment 39
Workflows that Run Mappings in a Hive Environment
You can add a mapping configured to run in a Hive environment to a Mapping task in a workflow. When you
deploy and run the workflow, the Mapping task runs the mapping.
You might want to run a mapping from a workflow so that you can run multiple mappings sequentially, make a
decision during the workflow, or send an email notifying users of the workflow status. Or, you can develop a
workflow that runs commands to perform steps before and after the mapping runs.
When a Mapping task runs a mapping configured to run in a Hive environment, do not assign the Mapping task
outputs to workflow variables. Mappings that run in a Hive environment do not provide the total number of target,
source, and error rows. When a Mapping task includes a mapping that runs in a Hive environment, the task
outputs contain a value of zero (0).
Configuring a Mapping to Run in a Hive Environment
You can use the Developer tool to configure a mapping to run in a Hive environment. To configure a mapping, you
must specify a Hive validation environment, a Hive run-time environment, and a Hive connection.
1. Open the mapping in the Developer tool.
2. In the Advanced properties, select Hive as the validation environment.
3. In the Run-time properties, select Hive as the run-time environment.
4. In the Run-time properties, select a Hive connection.
Hive Execution Plan
The Data Integration Service generates a Hive execution plan for a mapping when you run a mapping in a Hive
environment. A Hive execution plan is a series of Hive tasks that the Hive executor generates after it processes a
mapping for a Hive environment.
Hive Execution Plan Details
You can view the details of a Hive execution plan for a mapping from the Developer tool.
The following table describes the properties of a Hive execution plan:
Property Description
Script Name Name of the Hive script.
Script Hive script that the Data Integration Service generates based
on the mapping logic.
Depends On Tasks that the script depends on. Tasks include other scripts
and Data Integration Service tasks, like the Start task.
40 Chapter 5: Mappings in a Hive Environment
Viewing the Hive Execution Plan for a Mapping
You can view the Hive execution plan for a mapping that runs in a Hive environment. You do not have to run the
mapping to view the Hive execution plan in the Developer tool.
Note: You can also view the Hive execution plan in the Administrator tool.
1. In the Developer tool, open the mapping.
2. Select the Data Viewer tab.
3. Select Show Execution Plan.
The Data Viewer tab shows the the details for the Hive execution plan.
Monitoring a Mapping
You can monitor a mapping that is running on a Hadoop cluster.
1. Open the Monitoring tab in the Administrator tool.
2. Select Jobs in the Navigator.
3. Select the mapping job.
4. Click the View Logs for Selected Object button to view the run-time logs for the mapping.
The log shows the results of the Hive queries run by the Data Integration Service. This includes the location of
Hive session logs and Hive session history file.
5. To view the Hive execution plan for the mapping, select the Hive Query Plan view.
6. To view each script and query included in the Hive execution plan, expand the mapping job node, and select
the Hive script or query.
7. To view the MapReduce jobs in Jobtracker for a Hive query, select the query, and then click the job ID link in
the Properties view.
Jobtracker opens and shows details about MapReduce jobs that ran or are running on Hadoop.
Logs
The Data Integration Service generates log events when you run a mapping in a Hive environment.
You can view log events relating to different types of errors such as Hive connection failures, Hive query failures,
Hive command failures, or other Hadoop job failures. You can find the information about these log events in the
following log files:
LDTM log
The Logical DTM logs the results of the Hive queries run for the mapping. You can view the Logical DTM log
from the Developer tool or the Administrator tool for a mapping job.
Hive session log
For every Hive script in the Hive execution plan for a mapping, the Data Integration Service opens a Hive
session to run the Hive queries. A Hive session updates a log file in the following directory on the Data
Monitoring a Mapping 41
Integration Service node: <InformaticaInstallationDir>/tomcat/bin/disTemp/. The full path to the Hive
session log appears in the LDTM log.
Hadoop Log
To view the details about the MapReduce jobs for a Hive query, you can use the Hadoop JobTracker in the
Administrator tool to navigate to the Hadoop job page. You can also find the Hadoop JobTracker URL in the
LDTM log.
Troubleshooting a Mapping in a Hive Environment
When I run a mapping with a Hive source or a Hive target on a different cluster, the Data Integration Service fails to push the
mapping to Hadoop with the following error: Failed to execute query [exec0_query_6] with error code [10], error
message [FAILED: Error in semantic analysis: Line 1:181 Table not found customer_eur], and SQL state
[42000]].
When you run a mapping in a Hive environment, the Hive connection selected for the Hive source or Hive
target, and the mapping must be on the same Hive metastore.
42 Chapter 5: Mappings in a Hive Environment
C H A P T E R 6
Profiles
This chapter includes the following topics:
Profiles Overview, 43
Native and Hadoop Environments, 44
Profile Types on Hadoop, 46
Running a Single Data Object Profile on Hadoop, 47
Running Multiple Data Object Profiles on Hadoop, 48
Monitoring a Profile, 48
Viewing Profile Results, 49
Troubleshooting, 49
Profiles Overview
You can run a profile on HDFS and Hive data sources in the Hadoop environment. The Hadoop environment helps
improve the performance. The run-time environment, native Data Integration Service or Hadoop, does not affect
the profile results.
You can run a column profile, rule profile, and data domain discovery on a single data object profile in the Hadoop
environment. You can perform these profiling capabilities on both native and Hadoop data sources. A native data
source is a non-Hadoop source, such as a flat file, relational source, or mainframe source. A Hadoop data source
can be either a Hive or HDFS source.
If you use Informatica Developer, you can choose either native or Hadoop run-time environment to run a profile. If
you choose the Hadoop environment, the Developer tool sets the run-time environment in the profile definition.
Informatica Analyst supports native environment that uses the Data Integration Service.
You run a profile in the Hadoop run-time environment from the Developer tool. You validate a data source to run
the profile in both native and Hadoop environments. To validate the profile run in the Hadoop environment, you
must select a Hive connection. You can then choose to run the profile in either native or Hadoop run-time
environment.
You can view the Hive query plan in the Administrator tool. The Hive query plan consists of one or more scripts
that the Data Integration Service generates based on the logic defined in the profile. Each script contains Hive
queries that run against the Hive database. One query contains details about the MapReduce job. The remaining
queries perform other actions such as creating and dropping tables in the Hive database.
You can use the Monitoring tab of the Administrator tool to monitor a profile and Hive statements running on
Hadoop. You can expand a profile job to view the Hive queries generated for the profile. You can also view the run-
43
time log for each profile. The log shows run-time details, such as the time each task runs, the Hive queries that run
on Hadoop, and errors that occur.
The Monitoring tab contains the following views:
Properties view
The Properties view shows properties about the selected profile. You can access the MapReduce (MR)
details for the profile in Jobtracker from the Properties view. Jobtracker is a Hadoop component that shows
the status of MapReduce jobs that run on nodes in the Hadoop environment.
Hive Query Plan view
The Hive Query Plan view shows the Hive query plan for the selected profile.
Native and Hadoop Environments
When you run a profile in the native environment, the Analyst tool or Developer tool submits the profile jobs to the
Profiling Service Module. The Profiling Service Module then breaks down the profile jobs into a set of mappings.
The Data Integration Service runs these mappings and writes the profile results to the profile warehouse.
The native environment runs the mappings on the same machine where the Data Integration Service runs. The
Hadoop environment runs the mappings on a Hadoop cluster. The Data Integration Service pushes the mapping
execution to the Hadoop cluster through a Hive connection. This environment makes all the sources,
transformations, and Hive and HDFS sources available for profile run.
If you choose a native source for the Hadoop run-time environment, the Data Integration Service runs the profile
on Hadoop. You cannot run a Hadoop data source in the native run-time environment.
Supported Data Source and Run-time Environments
In the Developer tool, you can run a profile on native, Hive, and HDFS data sources. You can run a profile on both
Hive and HDFS sources in the Hadoop environment.
The following table describes the combination of data source types and run-time environments that Data Explorer
supports:
Data Source Type Run-time Environment
Native sources such as flat files, relational
sources, and mainframes
Native, Hadoop
Hive Hadoop
HDFS Hadoop
44 Chapter 6: Profiles
You cannot run some of the profile definitions in either the native or Hadoop environment. The following table
describes some of the run-time scenarios and whether you can run the profile in different run-time environments:
Scenario Hadoop Run-time Environment Native Run-time Environment
Running a profile on a Hive or HDFS
source within a mapping specification.
No No
Running a profile on a mapping
specification with a Hive or HDFS data
source.
Yes Yes
Running a profile on a logical data
object with a Hive or HDFS data source.
Yes Yes
Running a column profile on a mapping
or mapplet object with a Hive or Hadoop
source.
No Yes
Comparing the column profile results of
two objects in a mapping or mapplet
object with a Hive or HDFS source.
No Yes
Run-time Environment Setup and Validation
By default, all profiles run in the native run-time environment. You can change the run-time environment to
Hadoop in the Developer tool and run a profile. Before you run a profile, you need to verify whether the validation
settings in the profile definition match its run-time requirements.
The validation settings determine whether the profile definition suits the native run-time environment, Hadoop run-
time environment, or both. The steps to complete the run-time environment setup and validation are as follows:
1. Choose the validation environments. Validation environments are the environments that you want to set up for
the profile run. The Developer tool validates the data sources and transformations for these environments.
You must choose at least one of the environments. If you choose both environments, you must choose the
run-time environment for the profile.
2. Choose the run-time environment. When you choose the run-time environment, the Developer tool saves one
of the associated validation environments for profile run. If you choose Hadoop, you must select a Hive
connection. The Hive connection helps the Data Integration Service communicate with the Hadoop cluster to
push down the mapping execution from the Data Integration Service to the Hadoop cluster.
The validation environments determine whether the sources and transformations that any of the source rules and
data domains may contain are valid for the environments. The Developer tool validates a profile definition before
you run it.
Native and Hadoop Environments 45
The following table describes the validation environment settings that you can configure for a profile:
Option Description
Native (Data Integration Service) The Data Integration Service runs the profile.
Hadoop Runs the profile in the Hadoop environment. If you select this
option, you must specify the Hive connection.
Hive connection The Hive connection to run a profile in the Hadoop
environment.
You can specify both native and Hadoop options when you set up the validation environments for a profile. You
choose either Native or Hadoop as the run-time environment.
Run-time Environment and Profile Performance
In general, you run a profile on Hadoop data in the Hadoop run-time environment. For non-Hadoop data, profiles
on smaller data sources run faster in the native run-time environment.
You can run a profile on bigger data sources in the Hadoop run-time environment. In addition to the data size, you
also need to consider many other factors such as the network configuration, Data Integration Service
configuration, and Hadoop cluster configuration. Unless you need to run non-Hadoop data in the Hadoop run-time
environment at a later stage, you run a profile on data in the environment it resides.
Profile Types on Hadoop
You can run a column profile, data domain profile, and column profile with rules in the Hadoop environment.
You can run a column profile in the Hadoop environment to determine the characteristics of source columns such
as value frequency, percentages, patterns, and datatypes. Run a data domain profile in the Hadoop environment
to discover source column data that match predefined data domains based on data and column name rules. You
can also run a profile that has associated rules in the Hadoop environment.
Note: Random sampling may not apply when you run a column profile in the Hadoop environment.
Column Profiles on Hadoop
You can import a native or Hadoop data source into the Developer tool and then run a column profile on it. When
you create a column profile, you select the columns, set up filters, and sampling options. Column profile results
include value frequency distribution, unique values, null values, and datatypes.
Complete the following steps to run a column profile on Hadoop.
1. Open a connection in the Developer tool to import the native or Hadoop source.
2. Import the data source as a data object. The Developer tool saves the data object in the Model repository.
3. Create a profile on the imported data object.
4. Set up the configuration options. These options include validation environment settings, run-time settings, and
the Hive connection.
5. Run the profile to view the results.
46 Chapter 6: Profiles
Rule Profiles on Hadoop
You can run profiles on Hadoop that apply business rules to identify problems in the source data. In the Developer
tool, you can create a mapplet and validate the mapplet as a rule for reuse. You can also add a rule to a column
profile on Hadoop.
You cannot run profiles that contain stateful functions, such as MOVINGAVG, MOVINGSUM, DECODE or
COMPRESS.
For more information about stateful functions, see Rules and Guidelines for Functions.
Data Domain Discovery on Hadoop
Data domain discovery is the process of discovering logical datatypes in the data sources based on the semantics
of data. You can run a data domain profile on Hadoop and view the results in the Developer tool.
Data domain discovery results display statistics about columns that match data domains, including the percentage
of matching column data and whether column names match data domains. You can drill down the results further
for analysis, verify the results on all the rows of the data source, and add the results to a data model from the
profile model.
Running a Single Data Object Profile on Hadoop
After you set up the validation and run-time environments for a profile, you can run the profile to view its results.
1. In the Object Explorer view, select the data object you want to run a profile on.
2. Click File > New > Profile.
The profile wizard appears.
3. Select Profile and click Next.
4. Enter a name and description for the profile and verify the project location. If required, browse to a new
location.
Verify that Run Profile on finish is selected.
5. Click Next.
6. Configure the column profiling and domain discovery options.
7. Click Run Settings.
The Run Settings pane appears.
8. Select Hive as the validation environment.
You can select both Native and Hive as the validation environments.
9. Select Hive as the run-time environment.
10. Select a Hive connection.
11. Click Finish.
Running a Single Data Object Profile on Hadoop 47
Running Multiple Data Object Profiles on Hadoop
You can run a column profile on multiple data source objects. The Developer tool uses default column profiling
options to generate the results for multiple data sources.
1. In the Object Explorer view, select the data objects you want to run a profile on.
2. Click File > New > Profile to open the New Profile wizard.
3. Select Multiple Profiles and click Next.
4. Select the location where you want to create the profiles. You can create each profile at the same location of
the data object, or you can specify a common location for the profiles.
5. Verify that the names of the data objects you selected appear within the Data Objects section.
Optionally, click Add to add another data object.
6. Optionally, specify the number of rows to profile, and choose whether to run the profile when the wizard
completes.
7. Click Next.
The Run Settings pane appears. You can specify the Hive settings.
8. Select Hive and select a Hive connection.
You can select both Native and Hive as the validation environments.
9. In the Run-time Environment field, select Hive.
10. Click Finish.
11. Optionally, enter prefix and suffix strings to add to the profile names.
12. Click OK.
Monitoring a Profile
You can monitor a profile that is running on Hadoop.
1. Open the Monitoring tab in the Administrator tool.
2. Select Jobs in the Navigator.
3. Select the profiling job.
4. Click the View Logs for Selected Object button to view the run-time logs for the profile.
The log shows all the hive queries that the Data Integration Service ran on the Hadoop cluster.
5. To view the Hive query plan for the profile, select the Hive Query Plan view.
You can also view the Hive query plan in the Developer tool.
6. To view each script and query included in the Hive query plan, expand the profiling job node, and select the
Hive script or query.
7. To view the MapReduce jobs in Jobtracker for a Hive query, select the query, and then click the job ID link in
the Properties view.
Jobtracker opens and shows details about MapReduce jobs that ran or are running on Hadoop.
48 Chapter 6: Profiles
Viewing Profile Results
You can view the column profile and data domain discovery results after you run a profile on Hadoop.
1. In the Object Explorer view, select the profile you want to view the results for.
2. Right-click the profile and select Run Profile.
The Run Profile dialog box appears.
3. Click the Results tab, if not selected already, in the right pane.
You can view the column profile and data domain discovery results in separate panes.
Troubleshooting
Can I drill down on profile results if I run a profile in the Hadoop environment?
Yes, except for profiles in which you have set the option to drill down on staged data.
I get the following error message when I run a profile in the Hadoop environment: [LDTM_1055] The Integration Service failed to
generate a Hive workflow for mapping [Profile_CUSTOMER_INFO12_14258652520457390]." How do I resolve this?
This error can result from a data source, rule transformation, or run-time environment that is not supported in
the Hadoop environment. Refer Rules and Guidelines for Running Mappings in a Hadoop Environment for
more information on objects that are not valid in the Hadoop environment.
You can change the data source, rule, or run-time environment and run the profile again. View the profile log
file for more information on the error.
I see "N/A" in the profile results for all columns after I run a profile. How do I resolve this?
Verify that the profiling results are in the profiling warehouse. If you do not see the profile results, verify that
the database path is accurate in the HadoopEnv.properties file. You can also verify the database path from
the Hadoop job tracker on the Monitoring tab of the Administrator tool.
After I run a profile on a Hive source, I do not see the results. When I verify the Hadoop job tracker in the Administrator tool, I
see the following error when I open the profile job: "XML Parsing Error: no element found." What does this mean?
The Hive data source does not have any record and is empty. The data source must have a minimum of one
row of data for successful profile run.
After I run a profile on a Hive source, I cannot view some of the column patterns. Why?
When you import a Hive source, the Developer tool sets the precision for string columns to 4000. The
Developer tool cannot derive the pattern for a string column with a precision greater than 255. To resolve this
issue, set the precision of these string columns in the data source to 255 and run the profile again.
When I run a profile on large Hadoop sources, the profile job fails and I get an "execution failed" error. What can be the possible
cause?
One of the causes can be a connection issue. Perform the following steps to identify and resolve the
connection issue:
1. Go to the Monitoring tab in the Administrator tool.
2. Open the Hadoop job tracker.
3. Identify the profile job and open it to view the MapReduce jobs.
Viewing Profile Results 49
4. Click the hyperlink for the failed job to view the error message. If the error message contains the text
"java.net.ConnectException: Connection refused", the problem occured because of an issue with the
Hadoop cluster. Contact your network administrator to resolve the issue.
50 Chapter 6: Profiles
C H A P T E R 7
Native Environment Optimization
This chapter includes the following topics:
Native Environment Optimization Overview, 51
Processing Big Data on a Grid, 51
Processing Big Data on Partitions, 52
High Availability, 53
Native Environment Optimization Overview
You can optimize the native environment to increase performance. To increase performance, you can configure
the Integration Service to run on a grid and create partitions for PowerCenter sessions. You can also enable high
availability to ensure that the domain can continue running despite temporary network, hardware, or service
failures.
You can run profiles, sessions, and workflows on a grid to increase the processing bandwidth. A grid is an alias
assigned to a group of nodes that run profiles, sessions, and workflows. When you enable grid, the Integration
Service runs a service process on each available node of the grid to increase performance and scalability.
You can also run a PowerCenter session with partitioning to increase session performance. When you create
partitions for a PowerCenter session, the PowerCenter Integration Service performs the extract, transformation,
and load for each partition in parallel.
You can configure high availability for the domain. High availability eliminates a single point of failure in a domain
and provides minimal service interruption in the event of failure.
Processing Big Data on a Grid
You can run an Integration Service on a grid to increase the processing bandwidth. When you enable grid, the
Integration Service runs a service process on each available node of the grid to increase performance and
scalability.
Big data may require additional bandwidth to process large amounts of data. For example, when you run a Model
repository profile on an extremely large data set, the Data Integration Service grid splits the profile into multiple
mappings and runs the mappings simultaneously on different nodes in the grid.
51
Data Integration Service Grid
You can run Model repository mappings and profiles on a Data Integration Service grid.
When you run mappings on a grid, the Data Integration Service distributes the mappings to multiple DTM
processes on nodes in the grid. When you run a profile on a grid, the Data Integration Service splits the profile into
multiple mappings and distributes the mappings to multiple DTM processes on nodes in the grid.
For more information about the Data Integration Service grid, see the Informatica Administrator Guide.
PowerCenter Integration Service Grid
You can run PowerCenter repository sessions and workflows on a PowerCenter Integration Service grid.
When you run a session on a grid, the PowerCenter Integration Service distributes session threads to multiple
DTM processes on nodes in the grid. When you run a workflow on a grid, the PowerCenter Integration Service
distributes the workflow and tasks included in the workflow across the nodes in the grid.
For more information about the PowerCenter Integration Service grid, see the PowerCenter Advanced Workflow
Guide.
Grid Optimization
You can optimize the grid to increase performance and scalability of the Data Integration Service or PowerCenter
Integration Service.
To optimize the grid, complete the following tasks:
Add nodes to the grid.
Add nodes to the grid to increase processing bandwidth of the Integration Service.
Use a high-throughput network.
Use a high-throughput network when you access sources and targets over the network or when you run
PowerCenter sessions on a grid.
Store files in an optimal storage location for the PowerCenter Integration Service processes.
Store files on a shared file system when all of the PowerCenter Integration Service processes need to access
the files. You can store files on low-bandwidth and high-bandwidth shared file systems. Place files that are
accessed often on a high-bandwidth shared file system. Place files that are not accessed that often on a low-
bandwidth shared file system.
When only one PowerCenter Integration Service process has to access a file, store the file on the local
machine running the Integration Service process instead of a shared file system.
For more information, see the PowerCenter Performance Tuning Guide.
Processing Big Data on Partitions
You can run a PowerCenter session with partitioning to increase session performance. When you run a
PowerCenter session configured with partitioning, the PowerCenter Integration Service performs the extract,
transformation, and load for each partition in parallel.
For more information, see the PowerCenter Advanced Workflow Guide.
52 Chapter 7: Native Environment Optimization
Partition Optimization
You can optimize the partitioning of PowerCenter sessions to improve session performance. You can add more
partitions, select the best performing partition types, use more CPUs, and optimize the source or target database
for partitioning.
To optimize partitioning, perform the following tasks:
Increase the number of partitions.
Increase the number of partitions to enable the PowerCenter Integration Service to create multiple
connections to sources and process partitions of source data concurrently. Increasing the number of partitions
or partition points increases the number of threads, which also increases the load on the nodes in the
Integration Service. If the Integration Service node or nodes contain ample CPU bandwidth, processing rows
of data in a session concurrently can increase session performance.
Note: If you use a single-node Integration Service and you create a large number of partitions or partition
points in a session that processes large amounts of data, you can overload the system.
Select the best performing partition types at particular points in a pipeline.
Select the best performing partition type to optimize session performance. For example, use the database
partitioning partition type for source and target databases.
Use multiple CPUs.
If you have a symmetric multi-processing (SMP) platform, you can use multiple CPUs to concurrently process
session data or partitions of data.
Optimize the source database for partitioning.
You can optimize the source database for partitioning. For example, you can tune the database, enable
parallel queries, separate data into different table spaces, and group sorted data.
Optimize the target database for partitioning.
You can optimize the target database for partitioning. For example, you can enable parallel inserts into the
database and use a Router transformation to enable each partition to write to a single database partition.
For more information, see the PowerCenter Performance Tuning Guide.
High Availability
High availability eliminates a single point of failure in an Informatica domain and provides minimal service
interruption in the event of failure. When you configure high availability for a domain, the domain can continue
running despite temporary network, hardware, or service failures. You can configure high availability among the
Service Manager, PowerCenter and PowerExchange application services, PowerCenter Client, and command line
programs.
The following high availability components make services highly available in an Informatica domain:
Resilience. The ability of an Informatica domain to tolerate temporary connection failures until either the
resilience timeout expires or the failure is fixed.
Restart and failover. The restart of a service or task or the migration to a backup node after the service
becomes unavailable on the primary node.
Recovery. The completion of operations after a service is interrupted. After a service process restarts or fails
over, it restores the service state and recovers operations.
High Availability 53
When you plan a highly available Informatica environment, consider the differences between internal Informatica
components and systems that are external to Informatica. Internal components include the Service Manager,
application services, the PowerCenter Client, and command line programs. External systems include the network,
hardware, database management systems, FTP servers, message queues, and shared storage.
If you have the high availability option, you can achieve full high availability of internal Informatica components.
You can achieve high availability with external components based on the availability of those components. If you
do not have the high availability option, you can achieve some high availability of internal components.
Example
While you are fetching a mapping into the PowerCenter Designer workspace, the PowerCenter Repository Service
becomes unavailable, and the request fails. The PowerCenter Repository Service fails over to another node
because it cannot restart on the same node.
The PowerCenter Designer is resilient to temporary failures and tries to establish a connection to the PowerCenter
Repository Service. The PowerCenter Repository Service starts within the resilience timeout period, and the
PowerCenter Designer reestablishes the connection.
After the PowerCenter Designer reestablishes the connection, the PowerCenter Repository Service recovers from
the failed operation and fetches the mapping into the PowerCenter Designer workspace.
54 Chapter 7: Native Environment Optimization
A P P E N D I X A
Datatype Reference
This appendix includes the following topics:
Datatype Reference Overview, 55
Hive Complex Datatypes, 55
Hive Datatypes and Transformation Datatypes, 56
Datatype Reference Overview
Informatica Developer uses the following datatypes in Hive mappings:
Hive native datatypes. Hive datatypes appear in the physical data object column properties.
Transformation datatypes. Set of datatypes that appear in the transformations. They are internal datatypes
based on ANSI SQL-92 generic datatypes, which the Data Integration Service uses to move data across
platforms. Transformation datatypes appear in all transformations in a mapping.
When the Data Integration Service reads source data, it converts the native datatypes to the comparable
transformation datatypes before transforming the data. When the Data Integration Service writes to a target, it
converts the transformation datatypes to to the comparable native datatypes.
Hive Complex Datatypes
Hive complex datatypes such as arrays, maps, and structs are a composite of primitive or complex datatypes.
Informatica Developer represents the complex datatypes with the string dataype and uses delimiters to separate
the elements of the complex datatype.
Note: Hive complex datatypes in a Hive source or Hive target are not supported when you run mappings in a
Hadoop cluster.
55
The following table describes the transformation types and delimiters that are used to represent the complex
datatypes:
Complex Datatype Description
Array The elements in the array are of string datatype. Each element of the array is delimited by
commas. For example, an array of fruits is represented as [apple,banana,orange].
Map Maps contain key-value pairs and are represented as pairs of strings and integers
delimited by the = character. Each pair of string and integer pair is delimited by commas.
For example, a map of fruits is represented as [1=apple,2=banana,3=orange].
Struct Struct are represented as pairs of strings and integers delimited by the : character. Each
pair of string and integer pair is delimited by commas. For example, a map of fruits is
represented as [1,apple].
Hive Datatypes and Transformation Datatypes
The following table lists the Hive datatypes that Data Integration Service supports and the corresponding
transformation datatypes:
Hive Datatype Transformation Datatype Range and Description
Tiny Int Integer -32,768 to 32,767
Integer Integer -2,147,483,648 to 2,147,483,647 Precision 10, scale
0
Bigint Bigint -9,223,372,036,854,775,808 to
9,223,372,036,854,775,807 Precision 19, scale 0
Double Double Precision 15
Float Double Precision 15
String String 1 to 104,857,600 characters
Boolean* Integer 1 or 0
Arrays String 1 to 104,857,600 characters
Struct String 1 to 104,857,600 characters
Maps String 1 to 104,857,600 characters
* The default transformation type for boolean is integer. You can also set this to string datatype with values of True and False.
56 Appendix A: Datatype Reference
A P P E N D I X B
Glossary
A
Apache Hadoop
An open-source software framework that supports data-intensive distributed applications.
B
big data
A set of data that is so large and complex that it cannot be processed through standard database management
tools.
C
Cloudera's Distribution Including Apache Hadoop (CDH)
Cloudera's version of the open-source Hadoop software framework.
CompressionCodec
Hadoop compression interface. A codec is the implementation of a compression-decompression algorithm. In
Hadoop, a codec is represented by an implementation of the CompressionCodec interface.
D
DataNode
An HDFS node that stores data in the Hadoop File System. An HDFS cluster can have more than one DataNode,
with data replicated across them.
H
Hadoop cluster
A cluster of machines that is configured to run Hadoop applications and services. A typical Hadoop cluster
includes a master node and several worker nodes. The master node runs the master daemons JobTracker and
NameNode. A slave or worker node runs the DataNode and TaskTracker daemons. In small clusters, the master
node may also run the slave daemons.
Hadoop Distributed File System (HDFS)
A distributed file storage system used by Hadoop applications.
Hive environment
An environment that you can configure to run a mapping or a profile on a Hadoop Cluster. You must configure
Hive as the validation and run-time environment.
Hive
A data warehouse infrastructure built on top of Hadoop. Hive supports an SQL-like language called HiveQL for
data summarization, query, and analysis.
Hive executor
A component of the DTM that can simplify and convert a mapping or a profile to a Hive execution plan that runs on
a Hadoop cluster.
Hive execution plan
A series of Hive tasks that the Hive executor generates after it processes a mapping or a profile. A Hive execution
plan can also be referred to as a Hive workflow.
Hive scripts
Script in Hive query language that contain Hive queries and Hive commands to run the mapping.
Hive task
A task in the Hive execution plan. A Hive execution plan contains many Hive tasks. A Hive task contains a Hive
script.
J
JobTracker
A Hadoop service that coordinates map and reduce tasks and schedules them to run on TaskTrackers.
M
MapReduce
A programming model for processing large volumes of data in parallel.
MapReduce job
A unit of work that consists of the input data, the MapReduce program, and configuration information. Hadoop runs
the MapReduce job by dividing it into map tasks and reduce tasks.
metastore
A database that Hive uses to store metadata of the Hive tables stored in HDFS. Metastores can be local,
embedded, or remote.
N
NameNode
A node in the Hadoop cluster that manages the file system namespace, maintains the file system tree, and the
metadata for all the files and directories in the tree.
58 Glossary
native environment
The default environment in the Informatica domain that runs a mapping, a workflow, or a profile. The Integration
Service performs data extraction, transformation, and loading.
R
run-time environment
The environment you configure to run a mapping or a profile. The run-time environment can be native or Hive.
S
stateful variable port
A variable port that refers to values from previous rows.
T
TaskTracker
A node in the Hadoop cluster that runs tasks such as map or reduce tasks. TaskTrackers send progress reports to
the JobTracker.
V
validation environment
The environment you configure to validate a mapping or a profile. You validate a mapping or a profile to ensure
that it can run in a run-time environment. The validation environment can be Hive, native, or both.
Appendix B: Glossary 59
I NDEX
A
architecture
grid 3
Hive environment processing 4
native environment processing 3
B
big data
access 2
big data processing
example 5
C
column profiling on Hadoop
overview 46
connections
HDFS 20
Hive 20
D
data domain discovery on Hadoop
overview 47
Data Integration Service grid 52
Data Replication
description 2
installation and configuration 9
datatypes
Hive 56
Hive complex datatypes 55
G
grid
architecture 3
Data Integration Service 52
description 3, 51
optimization 52
PowerCenter Integration Service 52
H
HDFS connections
creating 24
properties 20
HDFS mappings
description 27
high availability
description 3, 53
Hive connections
creating 24
properties 21
Hive environment processing
architecture 4
Hive execution plan
description, for mapping 31
Hive mappings
description 28
workflows 40
Hive query
description, for mapping 31
Hive query plan
viewing, for mapping 41
viewing, for profile 48
Hive script
description, for mapping 31
I
Informatica adapters
installation and configuration 8
Informatica clients
installation and configuration 8
Informatica services
installation and configuration 8
M
mapping example
HDFS 27
Hive 29
Twitter 29
mapping run on Hadoop
monitoring 41
overview 31
N
native environment
high availability 53
mappings 26
optimization 51
Native environment processing
architecture 3
P
partitioning
description 3, 52
60
optimization 53
PowerCenter
installation and configuration 8
PowerCenter adapters
installation and configuration 8
PowerCenter Big Data Edition
overview 1
PowerCenter Integration Service grid 52
PowerCenter repository tasks
description 52
PowerCenter sessions
partitioning 52
PowerExchange adapters
installation and configuration 8
profile results
viewing 49
profile run on Hadoop
monitoring 48
Overview 43
profile types 46
running a single data object 47
running multiple data objects 48
R
rule profiling on Hadoop
overview 47
S
social media mappings
description 29
W
workflows
Hive mappings 40
Index 61