At this year's TechEd, Microsoft had several interesting Exchange-related announcements, so I wanted to summarize them in case you missed any of the news.
The biggest item on Microsoft's agenda was probably the announcement of Service Pack 2 (SP2) for Exchange 2010. This is a non-surprise surprise, given that SP2 had already been mentioned as shipping in the second half of 2011. Microsoft's Kevin Allison mentioned in his TEC 2011 keynote address that one of his key objectives is to get Exchange releases and updates into a regular rhythm, and this release fits into that pattern. Tony Redmond described SP2 as "not the headline release that SP1 was," and I think that's a fair assessment in most respects.
The new SP2 features, including a mobile-optimized version of Outlook Web App (OWA) called OWA Mini and support for address list segmentation, are interesting enough, but for my money the big changes in SP2 revolve around new tools for smoothing the process of delivering coexistence between on-premises Exchange 2010 installations and mailboxes. You should expect to see continued investment—and innovation—in this space. Microsoft is breaking new ground by delivering a single product that powers both its hosted service and on-premises mailboxes. As the Exchange and Office 365 teams gather operational experience and customer feedback, I expect them to move aggressively to smooth out any rough edges.
The biggest Exchange item out of TechEd for the rest of us was undoubtedly Microsoft's release of updated guidance for Exchange Server virtualization. Virtualized Exchange Unified Messaging (UM) servers are now supported, which is good news because a single server can already handle up to 100 or so concurrent calls. That's overkill for many sites, so being able to virtualize the UM role makes deploying it quite a bit more attractive.
More interestingly, there's new guidance surrounding the use of hypervisor-based disaster recovery technologies: Microsoft now supports mixing the use of Exchange high availability (in the form of database availability groups—DAGs) with hypervisor-based high availability, within limits. Here's the exact paragraph from Microsoft's support statement:
Exchange server virtual machines (including Exchange Mailbox virtual machines that are part of a DAG), may be combined with host-based failover clustering and migration technology, as long as the virtual machines are configured such that they will not save and restore state on disk when moved, or taken offline. All failover activity must result in a cold boot when the virtual machine is activated on the target node. All planned migration must either result in shutdown and cold boot, or an online migration that makes use of a technology like Hyper-V Live Migration.
This is carefully worded, and it's followed by a statement that the hypervisor vendor is responsible for providing support for such failover. The implication is pretty clear: Microsoft supports hypervisor-based high availability as long as the guest behaves like a physical machine would. You can't magically transport a failing server's workload over to another server without using clustering, and for now Microsoft doesn't support moving that workload over by moving a running virtual machine or by taking a point-in-time snapshot copy. Will the company support this approach in the future? The Magic 8-Ball says "signs unclear" for now.
Finally, in news unrelated to TechEd, Microsoft made a major announcement about Windows Phone 7 on 24 May. Among the new features they announced were a number of changes in the areas of Exchange integration, mobile device management, and security. I'll have more to say about those changes in a future column.