latoga labs

During the vSphere 4.1 beta period, two of my clients were very interested in the new Host Affinity Rules for issues they were having.  The day vSphere 4.1 was released I had a call to discuss Host Affinity Rules with another client to explore issues they were having that host affinity rules could solve.  Each of these global enterprise clients are running 1000’s of VMs in production and each have different uses for Host Affinity Rules that the average user may overlook.

I was planning on providing a quick overview of the Host Affinity Rules in this post, but Frank Denneman already provided a great overview that I can’t really improve on.  Start with his post to to understand the basics of the new Host Affinity Rules.

A few Host Affinity Rule use cases:

• Physical server based software licensing:  As hard as it is to believe, some software vendors still tie their software to physical computers.  Using Host Affinity rules allow you to purchase licenses for a subset of your physical servers in your vSphere cluster. (typically 2 so you have HA).  By forcing the VMs running the software in question to run on specific servers, you can ensure compliance with the software licensing.
• Isolation for troublesome VMs: as with most troubleshooting processes, the newest thing in an environment usually gets the blame.  For one of my clients this means that new workloads moved into the vSphere environment that have performance issues result in the application owners blaming vSphere for the performance issues.  Even after using something like vCenter AppSpeed to show the user where the performance problem exists in their application, app owners still won’t believe it until their VM is running by itself.  With Host Affinity Rules, you can force the problem-some VM to run on a server that has nothing else on it.  (Take that non-believer app owner!)
• Another approach to host pinning and reservations:  Some clients use pinning and reservations to help ensure certain levels of performance for some of their end users.  Like any over-riding control, you can eventually end up with more of these rules than can be realistically managed.  Host Affinity rules could be used in their place by forcing a more large grained control.  Of course, the same rule sprawl can occur with host affinity rules.  So it’s best to use them sparingly at first and really make sure the end users really need them before using them.  After all, cloud computing is supposed to make things easier for both end users and IT administrators…fight the gravitational pull of end user special needs.  More often than not these needs don’t really exist.

I’ve been discussing destkop virtualization a lot lately with customers, so when I can across this Dilbert cartoon it brough the paradigm shift caused by virtualized desktops to the forefront.  (More desktop virtualization discussion in the future…once I find the time to write again…)

As virtualization becomes more and more pervasive across the data center, many of my customers are now considering their vCenter Server as a tier 1 application.  This means more focus is being places on maintaining the availability of vCenter Server.  To quote Gene Kranz:

“Failure is not an option.”

vCenter Server is central to the following aspects of a virtualized data center:

• provides DRS & DPM monitoring and host load management
• enables vMotion (central to both DRS and DPM)
• centralized management portal for all VMs and ESX server running in a cluster (ESX and VMs still run without vCenter, but management become much more cumbersome)
• feeding data from VMs and ESX to other IT management platforms
• hosts SRM plugin for VM business continuity between data centers
• provisions desktops for View (desktop virtualization)

There are a number of different strategies that can be taken to provide availability of the vCenter Server, these typically fall into one of two categories: a cold standby server or a warm standby server.  Since the time needed to manually bring up a cold standby server for a large vSphere deployment can easily reach into the hours, most large organization tend toward a warm standby scenario and leverage some software automation to trigger the fail over.  There are many options here that fall into the general categories of clustering or host replication.  These tend to be complex and not always application specific.

To fill this gap and provide the monitoring and fail over needs of running vCenter Server as a teir 1 application, VMware recently released vCenter Server Heartbeat, which provides monitoring and automated fail over of both the vCenter Server and (optionally) the vCenter database.

Key aspects of Center Server Heartbeat:

• Monitors application (vCenter Server and optionally vCenter Database), network, and OS
  • underlying technology licensed from NeverFail for vCenter Server and SQL Server awareness and fail over
• Supports VM or Physical deployments of vCenter Server
• Uses replication engine to replicate data and transactions to standby server
• fail over of vCenter and Database across wan or LAN
• Protects from Split Brain scenario if a network outage were to occur
• Fail over of IP address so all hosts/VMs continue to function with vCenter normally
• Easy to configure, auto cloning of vCenter Server VMs (if deployed virtually) to create stand by server

vCenter Server Heartbeat Diagram

My recommended approach to providing Tier 1 availability of the vCenter server

• Ideally: run your vCenter server as a VM and utizliae vCenter Server Heartbeat to monitor and fail over vCenter.  All accomplished with the minimal amount of configuration due to vCenter Server Heartbeat’s VM cloning capabilities.
• Minimally: run vCenter server as a VM and configure a HA pair for that VM.  vCenter HA operates independantly of vCenter Server and will function even when the vCenter serer fails.  Becuase it is designed to provide general HA for a wide variety of situations, it is not application aware like vCenter Server Heartbeat.  Also, many architects don’t prefer this solution becuase the fail over is being provided by the tool that you are trying to protect.  But, it is better than no fail over solution for vCenter Server.

Here is a nice little Friday treat, VMware’s Srinivas Krishnamurti introduced vCenter Mobile Access today.  Were you ever on the road, or just out to lunch (physically) when you needed to check something on your VMware infrastructure?  With vCMA you would be able to do that from your mobile phone.  Check out the demo video:

vCMA will be available as a technology preview sometime in April. You will need to install a virtual appliance to run the vCMA server and connect it to your vCenter.

Yesterday I logged in to check out Yammer, the Twitter for Business.  What Yammer has created is the innovation that Twitter was not able to:  a way to create social media tool for businesses and come up with a business model to charge for the service.

Yammer is essentially a channel-ized twitter.  But the channel is only other people in your business; they force that by using the domain of your email address to create or add you to a channel.  Now, your status updates are only seen by your co-workers.  And if you start to carry on a conversation in your channel, you can view messages based on threads.

Where I think it gets interesting is their business model.  Offer the service for free, but then charge the company for an Admin privileges on the channel.  An Admin can brand the channel for the company, control members of the channel and even provide security by restricting the channel so you can only log in from the corporate network.

The problem with Yammer?  It’s yet another social communication channel.  The whole social networking services have become way to fractured.  Too many places to network.  Not enough time accomplishing anything.  To use the phrase “social not-working” is getting more an more applicable. (on that point, Yammer was developed by the Geni team…how’s that for not working.)  The advantage of a service like FriendFeed is that it is one place to check all your social networking feeds, even if you can only reply back via FriendFeed.

The openess of the web needs to be extended so that something like Yammer can be a piece of infrastructure that can be plugged into multiple other services.  One think I like about Twitter is that there are 3rd party interfaces.  I have enough web browser windows open on my desktop as it is, I need less not more.  As fredrickvan tweeted today, the key is keeping your social touch points in control.  While Yammer figured out a way to make money off of the status message, it’s just another social touch point that we have to manage.