No, the title of this post does not refer to a trip to the Mouse Kingdom.  And it’s not the lost Bob Hope/Bing Crosby movie.

Project Orlando is the code name for the next release of Citrix XenServer.  Companies don’t just use code names because we think it makes us sound mysterious.  Actually, it’s because it allows us marketing types to change the release numbers at the last minute for all sorts of arcane reasons, arousing the ire of engineers and release managers everywhere.

We are within a very short time of releasing the public beta of Orlando.  So between now and then, I’m going to write about one or two significant new capabilities or enhancements in XenServer that you’ll find in Orlando.  Then, on the big day, you’ll find the announcement here.

One more thing before I do, though: this time around, we’re going to be making the beta download available via the download section of  Instead of a separate download request form, you’ll need to log in with a My Citrix account.  If you already have one, you’re cool.  If you don’t, please create one, so you’ll be ready to rock the download when the software is available.

OK, that’s enough housekeeping.  Let’s get down to the goodies.

The first major enhancement to XenServer in Orlando is the availability of automated high availability (HA). The infrastructure of XenServer has offered the ability to script or manually manage availability, and the replicated configuration database has removed the potential single point of failure imposed by external management servers.  But customers have been looking for more automation, and here it is.

You will be able to take the virtual machines in a resource pool (on Platinum and Enterprise Editions) and identify whether you want the virtual machine to be restarted in the event that the server it’s running on fails.  You can even identify how high the priority for each one is, so in the case of multiple failures putting resources under stress, your most critical workloads will be returned to service.

We’ll also protect the master node of the resource pool, and if it fails, automatically designate another node as master -- no need for manual intervention there either.

In this release, you’ll need shared SAN storage to be available -- either Fibre Channel or iSCSI -- to be used in addition to a network connection as the “heartbeat” that determines if your servers are up.  (While it’s technically possible to store your VMs on NFS and to configure a separate small iSCSI or FC SR as the heartbeat disk, that approach can potentially cause issues if the connections to the VM storage fail while the heartbeat connection does not.)

The built-in HA capability isn’t your only option, of course.  Our partners will continue to provide solutions that also incorporate application-level protection, replication, remote protection, policy-based management, true continuous availability, and more.  But there will be a powerful HA capability that will meet the business continuity needs of most IT organizations right in the box.

(One other improvement will come along as a side-effect: the “automatic placement” capability -- start a VM on any available node -- will get smarter about which system is the best place to start a VM.)

And automated HA is just one of a list of new features and enhancements.  Check in tomorrow for the next stop on the road to Orlando,