Redundancy Architecting

Microsoft Exchange Server 2007 introduces several new features for high availability, as well as improvements to existing high availability features. In our Exchange solutions offerings, we make use of these new and improved features so that clients can pull off data and service availability for Exchange 2007 server roles. The new state of affairs enable organizations to separate high availability scenarios from site resilience scenarios, and to deploy configurations, that are tailored to the organisation's specific needs in each separate area.

New features for high availability and improvements to existing high availability features are available in Exchange 2007:

• Standby Continuous Replication (SCR)
• Multiple subnet failover clusters (Geo Clusters)
• Dynamic Host Configuration Protocol (DHCP) Internet Protocol version 4 (IPv4)
• IPv6
• Exchange and failover cluster network configuration
• New quorum models (disk and file share witness)
• Continuous replication (log shipping and seeding) over a redundant cluster network in a cluster continuous replication (CCR) environment
• Reporting and monitoring improvements
• Performance improvements
• Transport dumpster improvements
• Exchange Management Console improvements

Load Balancing

For high availability, you can load-balance Exchange front-end servers through Network Load Balancing (NLB). Utilising Microsoft’s native clustering technology, a farm of identically configured Exchange front end servers’ work together, uses an algorithm to provide load balancing and high availability.

Network Load Balancing is also architected for sharing user loads on web servers, active directory servers, ISA, VPN farms, Terminal Server farms etc. Once NLB is installed on primary server, more servers can be added to the NLB cluster. All the servers in the cluster are configured with two IPs - one hush-hush and the other which is shared by all the servers of one particular cluster. The algorithm decides turns in responding to client requests depending on the offered load and node availability. Under this, you can have a maximum of 32 servers in a cluster.

Clustering

With the advent of 2007 edition, Microsoft Exchange offers assorted clustering solutions to ensemble various business requirements and address concerns. Exchange Server Enterprise Edition supports clustering of a maximum of 4 nodes when using Windows 2000 Server and up to 8 nodes with Windows Server 2003.

Microsoft designed Exchange 2007 with built-in support for asynchronous replication identical to that of SQL servers’ log shipping for Cluster Continuous Replication (CCR) clusters which do not require shared external storage. This cluster is cost effective and can be either deployed in same location or extended across two locations as proper Exchange Disaster Recovery site.

The second type of cluster is Single Copy Cluster (SCC) which is the traditional clustering that was available in earlier editions of Exchange and this is one of the well-built architecture for higher number of users. On the cost effective front, there is Local Continuous Replication (LCR) which provides protection against local storage failures due to its model of data replication to alternative drives attached to the same physical server.

Finally there is Standby Continuous Replication (SCR) which enables the use of standby recovery servers.