There has been much talk about Microsoft Exchange 2016, so we thought we’ll show you some of the things that have changed from Microsoft Exchange 2013.
For starters, Microsoft Exchange 2016 is for the next-generation work environment, where much of the action happens in the cloud. This is why it comes with many innovative features that focus more on collaboration than mere communication, a practice that is being implemented across all organizations today. Also, there are some architectural changes for better speed, connectivity, and performance.
Let’s look at each of these features in depth.
Exchange 2016 uses a single building block architecture to make it easy to deploy, regardless of an organization’s size. It is also much simpler to set up and deploy than the previous versions, and this is sure to come as a huge relief to Exchange administrators. If you’ve already worked with Exchange in the past, the first thing you’ll notice at the time of setup is the missing CAS role option. This does not mean that CAS role is completely gone, rather, it is co-located to the back email server to enable multi-role exchange. In other words, the Mailbox server’s role is enhanced to include the logic needed to route protocol requests to their correct destination and also to have all the components needed for processing, storing, and rendering the data. Due to this modification, Exchange can make the best use of the underlying hardware.
You may wonder why this change was made and how it can benefit you. In the real world, there’s always a possibility for failures in discs and hardware. By removing CAS, Microsoft has reduced the server count, and at the same time, has increased the ability to sustain failures. Let’s say for example, you have four CAS and 10 mailbox servers for a total of 14 servers. If we assume that CAS servers play a dominant role in keeping the connections moving, then what happens when one of them fail? Obviously, the other three have to take up that increased load of the failed server, and this puts a big strain on them, even increasing their chances to fail soon.
To avoid this heavy load sharing, Exchange 2016 spreads the role of the failed server equally among all the remaining 13 servers, so the additional load on each server is relatively less. Further, all the servers perform the same role, so the number of servers needed to balance a load can be reduced to say 10 from 14. This way, the number of servers needed is reduced and the ability to recover from a failure is greatly increased.
In fact, this is one of the major changes you’ll see in Exchange 2016, and it’s in tune with the architectural changes that Microsoft has been making to place greater emphasis on simplifying architecture and improving their product’s capability.
With such a simplified architecture comes simplified migration. In fact, Exchange 2016 maybe the easiest version of Exchange that you’ll ever be able to migrate to, especially if you’re coming from Exchange 2010 or earlier. If you’re migrating, all that you’ll have to do is move your namespace to the new version, and that in turn will migrate the proxy to the old. In this sense, there will be really no change when you migrate.
If you’re already using Exchange 2013, then migration is all about adding an Exchange 2016 to your environment. There is no need to even move the namespace, as Exchange 2016 will proxy up from Exchange 2013!
The admin console for migration is super user friendly too. As an admin, all that you have to do is click on Migrations, and then click on “Move to a different database.” Next, pick the users you want to migrate, give a name to the migration batch, and click to finish it. Migration doesn’t get easier than this! You can also enter all the details and schedule the migration on a specific date. Since the entire migration is done online within just a few minutes, there is absolutely no impact on users. This way, migration is smooth and fast.
In Exchange 2016, Microsoft has made an attempt to build a self-healing system that would analyze data from hundreds of thousands of mailboxes, and based on it, predict, detect, and resolve failures. This automation is probably the best approach to handle failures, especially when you’re dealing with so many thousands of servers. Though Exchange 2013 had a limited set of tools called “Managed Availability,” it was mostly restricted to OWA or CAS servers providing OWA access to clients. In Exchange 2016, there are more well-developed tools that check the availability of all or individual Database Availability Group (DAG) at any time, based on which it’s easy to stay on top of bandwidth utilization and spot network issues.
One tool that needs a specific mention is the Database Diversion Detection tool that allows you to check logical corruptions occurring due to disc controller failures, hardware failures, and memory overflow. To identify these difficult failures, the tool calculates checksums for every single page on each database copy. This data is then sent to passive nodes that compare their checksums with the data received from the active nodes to detect anomalies. Automatically, you’ll get a notification of this discrepancy, so you can identify the problem and resolve it as soon as possible.
These tools are seen as the first step towards the process and technology of self-healing.
Search and connectivity
Search and connectivity are greatly improved in Exchange 2016. In the past, the search index was created on the active copy of the database, even across WANs. However, after taking inputs from many users, Microsoft has finally decided to create the index on the passive copy of the data, and this means the total cost of ownership is greatly reduced. Hopefully, this is great news for you!
Some changes have been made based on the analysis of past search patterns. So, this version’s search feature is optimized for short searches and misspelled words. Also, the responsiveness and speed of search queries has greatly improved in this version, both in OWA and in cache mode.
Next, to improve connection and reconnection speeds, Exchange 2016 has moved completely to MAPI over HTTP protocol, making MAPI over HTTP the default protocol for connectivity in Exchange 2016. However, if you use a client that doesn’t support MAPI, it will default to RPC over HTTP. Though the transition to move to MAPI started in Exchange 2013, it is fully implemented in this version.
You may wonder why this is such an important change. In past versions, Outlook disguised a MAPI request by encapsulating it in a RPC packed module, that in turn was wrapped into an HTTP request. Such complex packaging meant more overheads, and thus, longer connection times. In this version, the RPC encapsulation is removed from the HTTP packets, so the entire connection process is more simplified and faster. As a bonus, the HTTP headers will contain only meaningful data.
Moving toward the cloud
Exchange 2016 clearly signals Microsoft’s “cloud first” strategy, as this version is more cloud-oriented than the previous versions. It offers access to many cloud tools such as Office 365, SharePoint, and OneDrive to make document collaboration much easier than before. Also, the overall simplified architecture is more geared towards the cloud.
In short, Microsoft Exchange 2016 signals a new direction for Exchange servers and for Microsoft as a whole. Though there are no additional features in 2016 when compared to Exchange 2013, the changes in architecture and protocol can go a long way in improving the speed of connections. More importantly, this version lays the foundation for the future which is likely to be the cloud.