Sticking with the case study theme, one of Microsoft’s largest IT Consultancy Partners, Avanade, have suffered from an issue which businesses across the globe, big and small, can easily suffer from; server sprawl.  Their datacenter was full, causing energy and management costs to go through to roof but bring in Hyper-V R2 and System Center and they were able to reduce the physical footprint by 85%, reduce datacenter costs by 40%, and improve their database performance by 50%.  And yes, they’re virtualising SQL 2008, on Hyper-V.

Avanade_arch_Diagram

In fact, they’re running a 16-node, virtualised SQL Cluster across 2 Hyper-V clusters, running on the stonkingly-fast Intel E5500 and the also-very-fast Intel E5440 respectively.  One of these Hyper-V clusters (5500) is the High Performance one, and the other, is the ‘Standard’ performance cluster, but with Live Migration, they can move VMs within a cluster, yet with SCVMM, they can still shift the VM, and it’s backend storage, between the clusters, with the Quick Storage Migration capability.  Their total consolidation has taken them from 136 to 20 physical hosts, which doesn’t, on the surface, seem a massive ratio, but when you bear in mind the workloads they’re running, it’s still a significant achievement.  On the same Hyper-V Clusters, they also run Microsoft Dynamics CRM, Microsoft Office SharePoint Server 2007, Microsoft SharePoint Server 2010, Office Communications Server 2007, Microsoft BizTalk Server 2007, Microsoft Visual Studio Team Foundation Server, Microsoft System Center Operations Manager, and multiple Avanade-authored applications. The Avanade-authored applications include a time-entry system; a contact management system; an extract, transform, and load (ETL) system that processes data going into a data warehouse; a learning management system; and a knowledge management repository.

“Unexpected quote of the case study” has to be:

“For example, it takes an hour to run SQL Server Integration Services on a physical server, but it takes 30 minutes to run it on a virtual machine—that’s on top of supporting many other applications on the server. The performance of key applications, such as Microsoft Dynamics CRM, has also improved dramatically. This benefit of improved database performance is that our users can get their hands on reports in half the time required before.”

Sure, that will be down to the fact that it’s running virtualised on something like the 5500 under the covers, but still, it’s a massive leap from the ‘what hit am I going to take by virtualising’ perception that can still plague the industry, no matter how hard we, VMware or Citrix try to convince people otherwise!

You can read the Microsoft Case Study for this deployment, here.

So, that’s the server backend, and the virtualisation platform, but that doesn’t run itself, without decent storage, and when I think decent storage, I lean towards NetApp.  Avanade tend to agree…

If you read the Microsoft case study, linked to above, you’ll see that the virtual infrastructure shares one NetApp FAS 3040 SAN, with 50TB of storage, however, they decided to upgrade this to a FAS3170HA for even greater scalability.  Our case study doesn’t go into any further detail than that, however if you head on over to the NetApp case study, for the same deployment, you’ll satisfy your hunger for storage-related-goodness!

Also in the NetApp case study, you’ll see that the number of VM’s have increased, now up to 350 on Hyper-V R2, which is actually more inline with what Patrick Cimprich, from Avanade details in his article – there are also some slight differences in the hardware detailed for the hosts between the 2 case studies:

AvanadeCaseNetApp

(Click image to enlarge)

Either way, it’s a powerful infrastructure, and is letting them consolidate significantly, whilst boosting performance.  One of the big features that I frequently talk about as a winner with Partners/Customers, is the DeDuplication, and it’s something that Patrick has also called out:

Combine CSVs with NetApp deduplication. Since re-architecting our NetApp aggregates and volumes for cluster shared volumes, we’ve noticed a big improvement in capacity savings by using NetApp deduplication. We are seeing an average deduplication ratio of 50% across our entire virtualization storage infrastructure. Given the size of our VM environment before deduplication (around 7 to 8TB), that’s been a significant savings of 3 to 4TB, which supports our plans for adding many more VMs. Our use of the NetApp system with CSVs provided the right combination to really let deduplication shine.

From the image above, you’ll also see the use of System Center.  From Virtual Machine Manager, to Operations Manager, and on to Data Protection Manager (where’s SCCM!  Must be just off the edge of the slide ;-)), the System Center technologies are providing Avanade with a centrally-well-managed infrastructure.  The integration with NetApp, from a management perspective, is a fantastic value add to the infrastructure, whether it’s through SnapManager for Hyper-V, or, more specifically, the integration between NetApp storage, and System Center Operations Manager, with ApplianceWatch PRO 2.0.

If you’re not familiar with ApplianceWatch PRO 2.0, and what it gives you, read on…

Firstly, it’s free (we like).  Secondly, you download it, import it into System Center Operations Manager, and you will start to see these benefits:

  • Use Microsoft technology to simplify and monitor all of your NetApp storage systems
  • Drill down into NetApp storage systems to view your hardware and storage layout
  • Isolate problems quickly using Microsoft Operations Manager’s alert and health explorer views.
  • Configure deduplication and autosize settings on NetApp flexible volumes

Is that it?  Nope!  If you’ve read my post the other day, around IBM, Brocade and System Center, you should have a handle on what PRO is all about, and those eagle-eyed among you, will have noticed that the ApplianceWatch MP, is a PRO MP, so brings about the following benefits in a virtual environment:

  • Monitor storage availability and capacity as they pertain to virtual machines running on a Microsoft Hyper-V host.
  • Receive PRO tips with auto-remediation support within the System Center Virtual Machine Manager 2008 R2 console

Cool stuff hey?  You can read more about some of the specific PRO Tips that can be generated in SCVMM, over on the NetApp MSEnviro blog.  Add to that, the specific guidance that NetApp have produced around Storage Best Practices for Microsoft Virtualisation, and you’ve got plenty of weekend reading ahead of you!