Scale Computing for Exchange 2010

Microsoft has added some important new features to Microsoft Exchange 2010 that play perfectly into Scale Computing’s storage offerings. Combined, they will provide a very robust, high performance, highly available solution for companies of all sizes.

If you are familiar with Exchange 2007, you know the different clustering solutions offered. As in Exchange 2000 and 2003, Exchange 2007 offered SCC, or single copy cluster. This was the traditional Exchange clustering solution which used shared storage to host the Exchange database, transaction logs and also the Windows cluster quorum. New to Exchange 2007 was LCR (local continuous replication), which was used by small businesses to replicate a copy of their Exchange DB to a separate disk on the same server, CCR (cluster continuous replication), used to replicate Exchange DB’s between two servers for storage redundancy, and SCR (standby continuous replication), which could be used to replicate Exchange DB’s to a disaster recovery site.

With Exchange 2010, all of the different cluster solutions found in 2007 have been replaced. You now have what are called DAGs, or Database Availability Groups. A DAG contains multiple Exchange mailbox servers as members. A database is associated with one or multiple servers within a DAG. The same DB copy can live on up to 16 different servers within a DAG. This new clustering technology partners well with Scale’s low cost storage, enabling an affordable disaster recovery solution.

With Scale Computing’s ICS storage technology, all of the Microsoft Exchange mailbox servers in your organization can utilize iSCSI based LUNS in a centralized, highly available, high performance SAN environment. Not only is there redundancy at the server level, with multiple mailbox servers within a single DAG, there is also the redundancy built into the Scale ICS cluster, at both the disk and storage node level.

As data demands increase, simply add additional nodes, which not only increases capacity but also bandwidth and performance, with zero downtime. The cluster automatically detects the added node and grows the storage in a redundant manner, with no interruption to service. This zero downtime scalability is crucial for systems administrators that have trouble finding downtime and change windows for their production, 24/7 storage infrastructure.

When looking at the clustering/replication architecture that Exchange 2010 is now based on, Scale Computing offers a very compelling storage solution. Microsoft has moved away from needing to use overpriced, overly complicated storage systems to back-end their Exchange Databases and Transaction logs. With transaction log shipping to seed and replicate copies of DB’s on separate servers and disks, and the new node majority and file share witness that replaces a shared quorum disk, complicated and cumbersome shared disk clustering is gone with Exchange 2010. With no single point of failure, high performance and zero downtime scalability, combined with the lowest entry point on the market, Scale Computing storage combined with Microsoft Exchange 2010 is a winning combination.