Monday, February 1, 2010

High Availability | From Exchange 2007 to Exchange 2010

High Availability is one of the most important factors considered today in almost all the messaging deployments as email is becoming the mission critcal applcation and the backbone for all the businesses.

Exchange has come a long way from its earlier versions - Exchange 2003 that uses typical Windows Clustering technologies based on a shared storage model, and then to Exchange 2007 that brought new dimensions with the introduction of log replication technology in LCR, CCR and SCR. With this new technology of continuos replication where the transaction logs are shipped from one copy of a database to another, the exchange 2007 deployment offers high availability in
various scenarios like Local Continuous Replication (LCR) on a single server deployment, Cluster Continuous Replication (CCR) available accross different servers and Standy Continuous Replication (SCR) spread across different sites.

Exchange 2010, introduces the concept of Database Availability Groups (DAG) that takes the high availability to the next level for mailbox servers. A DAG is as the name suggests a group or a collection of mailbox servers (upto a maximum of 16) that uses the continuous replication technology that was first introduced in Exchange 2007 and are effectively a combination of Cluster Continuous Replication (CCR) and Standby Continuous Replication (SCR). It also makes use of
some of the components of Windows Failover Clustering to achieve high availbility and these cluster elements are installed automatically when a mailbox server is added to a DAG and managed completely by Exchange.

To achieve full high availibility solution for all the roles, with the introduction of Exchange Server 2007 SP1, it can be achieved by deplyoing a total of minimum of 4 servers - two servers installed as a single CCR environment, giving high availability for the users’ mailboxes and the other two servers deployed as combined Hub Transport and Client Access Servers, and configured as a load-balanced pair.

But with Exchange 2010, a full high availability solution can now be deployed by using a minimum of 2 servers as it’s now possible to combine the mailbox server role with other roles such as the Hub Transport and Client Access Server role. And with futher IO reductions in Exchange 2010 and RAID-less/JBOD support, it support much larger mailboxes with reduced storage costs.

To summarize Exchange 2010 high availibilty improvements:

  • Combines the capabilities of CCR and SCR into one platform.
  • Easier than traditional clustering to deploy and manage.
  • Allows each database to have up to 16 replicated copies.
  • Provides full redundancy of Exchange roles on two servers.
  • Further IO reductions.
  • RAID-less / JBOD support.
So exciting times ahead for the folks who are planning migration to Exchange 2010 :)