Windows Server 2008 R2 provides two clustering technologies, which are both included on the Enterprise and Datacenter Editions. Clustering is the grouping of independent server nodes that are accessed and viewed on the network as a single system. When a service and/or application is run from a cluster, the end user can connect to a single cluster node to perform his work, or each request can be handled by multiple nodes in the cluster. In cases where data is read-only, the client might request data from one server in the cluster and the next request might be made to a different server and the client would never know the difference. Also, if a single node on a multiple node cluster fails, the remaining nodes will continue to service client requests and only the clients that were originally connected to the failed node may notice either a slight interruption in service, or their entire session might need to be restarted depending on the service or application in use and the particular clustering technology that is in use for that cluster.
The first clustering technology provided with Windows Server 2008 R2, Enterprise and Datacenter Editions is failover clustering. Failover clusters provide system fault tolerance through a process called failover. When a system or node in the cluster fails or is unable to respond to client requests, the clustered services or applications that were running on that particular node are taken offline and moved to another available node where functionality and access are restored. Failover clusters, in most deployments, require access to shared data storage and are best suited, but not necessarily limited to, the deployment of the following services and applications:
» File services—File services deployed on failover clusters provide much of the same functionality a standalone Windows Server 2008 R2 system can provide, but when deployed as clustered file services, a single data storage repository can be presented and accessed by clients through the currently assigned and available cluster node without replicating the file data.
» Print services—Print services deployed on failover clusters have one main advantage over a standalone print server: If the active print server fails, each of the shared printers is made available to clients using another designated print server in the cluster. Although deploying and replacing printers to computers and users is easily managed using Group Policy deployed printers, when standalone print servers fail, the impact can be huge, especially when servers, devices, services, and applications that cannot be managed with group policies access these printers.
» Database services—When large organizations deploy line-of-business applications, e-commerce, or any other critical services or applications that require a back-end database system that must be highly available, deploying database services on failover clusters is the preferred method. Also, in many cases configuring enterprise database services can take hours and the size of the databases, indexes, and logs can be huge, so deploying database services on a standalone system encountering a system failure may results in several hours of undesired downtime during repair or restore, instead of quick recovery as with a failover cluster.
» Back-end enterprise messaging systems—For many of the same reasons as cited previously for deploying database services, enterprise messaging services have become critical to many organizations and are best deployed in failover clusters.
» Hyper-V virtual machines—As many organizations move toward server consolidation and conversion of physical servers to virtual servers, providing a means to also maintain high availability and reliability has become even more essential when a single physical Hyper-V host has several critical virtual machines running on it.
The second Windows Server 2008 R2 clustering technology is Network Load Balancing (NLB), which is best suited to provide fault tolerance for front-end web applications and websites, Remote Desktop Services Session Host server systems, VPN servers, streaming media servers, and proxy servers. NLB provides fault tolerance by having each server in the cluster individually run the network services or applications, removing any single points of failure. Depending on the particular needs of the service or application deployed on an NLB cluster, there are different configuration or affinity options to determine how clients will be connected to the back-end NLB cluster nodes. For example, on a read-only website, client requests can be directed to any of the NLB cluster nodes; during a single visit to a website, a client might be connected to different NLB cluster nodes. As another example, when a client attempts to utilize an e-commerce application to purchase goods or services provided through a web-based application on an NLB cluster, the client session should be initiated and serviced by a single node in the cluster, as this session will most likely be using Secure Sockets Layer (SSL) encryption and will also contain specific session data, including the contents of the shopping cart and the end-user specific information.
Source of Information : Sams - Windows Server 2008 R2 Unleashed (2010)