Anda di halaman 1dari 10

CLOUD COMPUTING

ARCHITECHURE
MANAGING BUSINESS ON CLOUD
HYPERVISOR FAILOVER CLUSTERING

ISM634
• A failover cluster is a group of independent
computers that work together to increase the
availability and scalability of clustered roles.

• If one or more of the cluster nodes fail, other


nodes begin to provide service (a process
known as failover).

• The failure of a hypervisor or its underlying


physical server cascades to all hosted virtual GROUP-1
servers further causing their hosted IT
resources to fail. TEAM MEMBERS:
containers ASHISH SHARMA
• Hypervisors are clustered across multiple
physical servers, so that if one fails, active AKASH SHARMA
virtual servers are transferred to another
node.
JASLEEN KAUR
Fig -Two Virtual severs experience failure after their host physical server goes down
ESHITA GULAIYA
• Heartbeat messages are passed between
clustered hypervisors and a central VIM to HARSHA BHARDWAJ
maintain status monitoring. HEENA DUA
WORKING

Fig :-A cloud architecture resulting from the application of the Hypervisor Clustering Fig :-A cloud architecture resulting from the application of the Hypervisor Clustering
FAILOVER SYSTEM HEALTH MONITERING OF NODES IN CLUSTER

• Failover clustering conducts health monitoring


between nodes to detect when nodes are no
longer available.
• When nodes are unresponsive clustering
takes recovery action.
• Unicast in nature and uses a Request-Reply
process for reliability.

• Roles are running on a particular node, and a


heartbeat signal is exchanged between the
nodes.

• Once the Node becomes unresponsive, roles


are transferred to the other node in the
cluster, without affecting the users.
CLUSTER COMMUNICATION CLUSTERING SCALE
• A single cluster can have up-to 64 nodes.

• A single cluster can host up-to 8000 VM’s.

NETWORK HEALTH DETECTION


NETWORK HEALTH INTRA-CLUSTER CSV REDIRECTION
MONITORING COMMUNICATION
• All I/O failure • Nodes exchange heartbeat every 1 second
• Heartbeats are • Database updates scenarios.
and state (configurable).
sent to monitor • Sent over same
health status of synchronization • Nodes are considered down, if they do not
network as intra-
network that are sent respond to 5 heartbeats. (Configurable).
cluster.
interfaces. between the • Nodes are removed from cluster
• Are sent over all nodes in the membership if they exceed threshold.
cluster enabled cluster.
networks. • E.g. When creating
a new resource
the cluster
database must be
updated on all
nodes.
Hypervisor clustering cloud service provider ( EC3) GENERAL ARCHITECHURE
• For the first step, a launcher (Elastic Cloud Computing
Cluster Launcher or EC3-L) has been developed that
deploys the front-end on the Cloud using the infrastructure
deployment services

• The sysadmin will run this tool , providing it with the


following information

Maximum cluster size :-Establishes a cost limit in case of a


workload peak. The maximum cluster size can be modified at
any time once the virtual cluster is operating. Thus, the
sysadmins can adapt the maximum cluster size to the dynamic
requirements of their users. In this case the LRMS must be
EC3: Elastic Cloud Computing Cluster
reconfigured to add the new set of virtual nodes and in some
cases it may imply a LRMS service restart.
• It is a tool to create elastic virtual clusters on
top of Infrastructure as a Service (IaaS) RADL document specifying the desired features of the cluster
providers, either public (such as Amazon front-end, regarding both hardware and software (OS, LRMS,
Web Services, Google Cloud or Microsoft additional libraries, etc.). These requirements are taken by the
Azure) or on-premises (such as OpenNebula launcher and extended to include additional ones (such as
and Open Stack) installing CLUES and its requirements together with the
• The user is provided with a remote secure libraries employed to interact with the IaaS Cloud provider,
shell as the entry point to the cluster. etc.) in order to manage elasticity.
Direct LUN Access Architecture

• In Direct LUN Access architecture, virtual servers are


allowed to directly access the RAW block-based
storage via physical HBA card
Cloud Usage Monitor
• Effective approach as virtual servers in the same
cluster can use the LUN as shared volume for
clustered databases Pay-Per-Use Usage
Monitor
• Virtual Servers physical Connectivity to the LUN and
cloud storage device is enabled by Physical Hosts

• LUN’s are created and configured on the cloud storage Resource Replication
device for LUN presentation to the hypervisors

• Cloud storage device needs to be configured using raw


device mapping to make LUN’s visible to the virtual Mechanisms which can be
server as block based RAW SAN LUN incorporated into this
Architecture
• LUN needs to represented with a unique LUN ID to be
used by all of the virtual servers as shared storage
Fig 1:-Hypervisor shows the mapped LUN’s to the Fig 2:-Hypervisor(2) maps the presented LUN’s Fig 3:-Virtual Server’s Storage commands are
virtual servers as normal file based storage to be directly to the virtual servers(3). Virtual Servers can received by the hypervisors(5),which process
used as such(3) see the LUN as RAW Block based Storage & can and forward the requests to the storage
access them directly(4). processor(6)
HIGH AVAILABILITY

• Refers to a system or component that is continuously


operational for a desirably long length of time.
• Availability can be measured relative to "100%
operational" or "never failing."
• A widely-held but difficult-to-achieve standard of
availability for a system or product is known as "five
9s" (99.999 percent) availability.

HIGH AVAILABILITY THROUGHOUT DATACENTER


• HARDWARE HIGH AVAILABILITY (Servers, storage,
network).
• WORKLOAD HA (Host Failover clustering for VM’s).
• MANAGEMENT HA (System Center).
• SITE HA (Replica, Multi-Site Clustering).

WHY HIGH AVAILABILITY?


• Protection from downtime.
• Simplified Maintenance activities.
• Improves resilience and agility.
• Flexibility.
HOW AMAZON DO IT
Amazon EBS
Amazon EBS provides durable, block-level storage
volumes that you can attach to a running instance.

Amazon EC2 Instance Store


Many instances can access storage from disks that
are physically attached to the host computer.
Instance store provides temporary block-level
storage for instances.
Amazon EFS File System
Amazon EFS provides scalable file storage for use
with Amazon EC2.
Amazon S3
Amazon S3 provides access to reliable and
inexpensive data storage infrastructure. It is
designed to make web-scale computing easier by
enabling you to store and retrieve any amount of
data, at any time, from within Amazon EC2 or
anywhere on the web.

Anda mungkin juga menyukai