Hot points from Exchange Connections

I don't know about you, but I seem to spend every waking hour talking about technology at conferences. It's no different at Exchange Connections, taking place in Sin City this week. Not much sin to be seen, possibly because we're heads-down discussing Exchange... Here are some of the major talking points that I've heard from attendees so far.

It’s been very satisfying to see such a great bunch of people turn up for Exchange Connections, which truly lived up to its ambition of being the best independent conference covering Exchange. It's not only the speakers, of course, because they are always great (I have to say that otherwise they’ll never submit a session proposal again). The important thing is that lots of attendees are here asking some great questions about Exchange 2010, Exchange 2013, Office 365, hybrid connectivity, ActiveSync, and Lync interoperability.

Talking to attendees at breaks and after sessions, here are some of the hot points that I’ve picked up so far:

  • The quality of the updates issued by Microsoft for Exchange 2013 and the overall product quality in the initial release is concerning many attendees, especially because customers feel under pressure to rush cumulative updates into production “too quickly”, if only to make sure that their environment remains supported. The point being made is that if Microsoft cannot be trusted to deliver quality code, how can companies protect themselves when they assemble all the other bits around Exchange to build production environments?
  • Most of the companies represented at Connections seem to be running Exchange 2007 or Exchange 2010. As expected, far fewer run Exchange 2003 than last year while relatively few have moved to Exchange 2013 to date. Upgrading to newer versions of Exchange has led to larger mailboxes with about 20% reporting that they use 10GB or larger quotas. Another interesting development is that roughly 10% of companies now run Exchange Unified Messaging, so UM is clearly beginning to be more popular.
  • Microsoft’s rush to embrace the cloud is seen as both good and bad. It’s good when it forces positive developments in the product, like the ever-decreasing I/O footprint for Exchange and the way that features are tested in Office 365 before being released to on-premises customers. It’s not so good when that testing fails to happen.  The reduction in I/O requirements means that more companies are deploying Exchange 2010 and Exchange 2013 with DAS/JBOD-based storage, something that delighted Jeff Mealiffe when he did a straw poll at his performance and sizing session on Tuesday.
  • Being forced to deploy a new version of Outlook to gain access to new Exchange features is an age-old problem. Even though Greg Taylor of Microsoft told me that he thinks Outlook 2013's OST “slider” control is great, the situation is no different with Exchange 2013 and Outlook 2013. It seems though that most companies are happy to go ahead and use Exchange 2013 alongside with Outlook 2007 or Outlook 2010 as these clients are “good enough”. Both clients support Outlook Anywhere, which is the fundamental thing for connection to Exchange 2013, and the features that will be missing (like Data Loss Prevention in Outlook 2013 or MailTips in Outlook 2010) are not deemed sufficient to justify touching so many user desktops. On the other hand, companies moving from Windows XP seem to be planning complete refreshes and are including Office 2013 in the mix.
  • The appearance of new iOS devices and iOS 7 has caused some concern, especially for those who allow an open BYOD policy. Find out more in this post. Perhaps the thought of more ActiveSync bugs in a new version of Apple’s mail app will cause people to consider using Microsoft’s OWA for Devices app (that is, when it supports on-premises Exchange - the word is, apparently, "soon").
  • The impact of PRISM is being felt in terms of projects slowing down. Those who have already committed to Office 365 are going there, but perhaps a little more slowly as people digest just what PRISM means to their company. Those who have not signed up for Office 365 seem to be putting that decision off more than they were before and are considering other alternatives, including staying on-premises until the fog of confusion clears. The hard stop effect of PRISM is even more obvious outside the U.S. where non-U.S. companies are looking for hosted alternatives offered by local companies.

The most interesting story was one reported of someone (who shall stay anonymous) whose management decided that the number of database failovers is a rather important metric and that their Exchange administrators should be judged on keeping failovers to a minimum. He wasn't impressed when he discovered that Exchange 2013's Managed Availability automated monitoring and problem resolution framework could failover databases if probes reported that a server's health set was not optimal (this EHLO blog post is a good one to read to understand why). Automated failovers are all very well as they keep databases online and users happy, but they can really screw up a stupid management metric!

Of course, companies vary dramatically in terms of environment, needs, and capabilities and some or all of the points mentioned above might not affect you. But I bet that some do. And if you are at Connections and have strong feelings on any of the above (or other topics), why not join the debate by coming along to see The UC Architects take center stage for a live session at 1pm on Wednesday. Given the make-up of The UC Architects, I doubt that this bunch will hold back on anything. It should be an interesting 90 minutes.

Follow Tony @12Knocksinna

Discuss this Blog Entry 5

on Oct 3, 2013

Thanks for the analysis, Tony,

I think the uptick in UM deployments is attributable to the surge in Lync deployments. Lyn and Exchange UM go together like two peas in a pod. I see companies replacing their old PBX's with Lync and then integrating it with Exchange UM to lower costs and improve functionality.

Regarding I/O, it's becoming basically irrelevant. When JBOD can provide the number of IOPS required by Exchange, it becomes less important as a design factor. Don't get me wrong - it's still critically important to JetStress your storage subsystem to know it will function properly for a given design. It's just less likely that it will fail to deliver enough IOPS or fail due to latency.

Congratulations on chairing a great conference!

on Oct 4, 2013

Thank you for your high lights! Are those sessions available somewhere for us who have been unable to participant?

About the attendee's concerns: "....especially because customers feel under pressure to rush cumulative updates into production “too quickly”...." I feel this is just a method to grill customers to go into O365 "someone else do the updates for you".

on Oct 7, 2013

Some IT/Dev Connections sessions are available in our store at, and more are on the way. ---Amy Eisenberg, Editor-in-Chief, Windows IT Pro

on Oct 5, 2013

@Pepi_1. I think that the conference attendees get the content but that the sessions are not posted for general access. Sorry, but the attendees have to get value for their payment!


on Oct 7, 2013

@Pepi_1, in addition to Amy's comment, I also know that some of the Exchange Connections speakers have posted their decks on SlideShare.

Please or Register to post comments.

What's Tony Redmond's Exchange Unwashed Blog?

On-premises and cloud-based Microsoft Exchange Server and all the associated technology that runs alongside Microsoft's enterprise messaging server.


Tony Redmond

Tony Redmond is a senior contributing editor for Windows IT Pro. His latest books are Office 365 for Exchange Professionals (eBook, May 2015) and Microsoft Exchange Server 2013 Inside Out: Mailbox...
Blog Archive

Sponsored Introduction Continue on to (or wait seconds) ×