VMware released VMware vSphere 4 with much fanfare earlier this year and spent the rest of the year building out...
the rest of its product suite to take advantage of vSphere. VMware has been pushing customers to upgrade, offering special discounts for upgrades to the high-end Enterprise Plus licensing tier and enticing customers with new features like VMware Fault Tolerance (FT), larger CPU and memory limits for hosts and VMs and distributed network switches. Most recently, VMware released Update 1 for VMware vSphere 4.0. The release of the first update is seen by many as a catalyst for increased adoption of the product, as organizations hesitant to adopt "dot zero" releases may now move forward.
Should you upgrade your VMware Infrastructure 3 (VI3) environment to VMware vSphere 4 or should you wait? There are two aspects to consider when trying to answer this question: the technical aspect and the business aspect.
Evaluating the upgrade from a technical perspective
Most users who have taken a look at VMware vSphere 4.0 from a technical perspective agree that the new version offers some significant technical reasons for upgrading. These features have been covered in great detail by numerous other authors so I won't repeat that material here, but some of these new features include:
- Greatly expanded resource limits for ESX/ESXi hosts and virtual machines
- vNetwork Distributed Switch (vDS)
- Support for third-party distributed switches, such as the Nexus 1000V
- VMware Fault Tolerance (FT)
- Host profiles
- Support for third-party multipathing modules
- vCenter Server linked mode groups
- Expanded limits on the number of hosts and guests vCenter Server can manage
- Support for new guest operating systems (such as the Windows 7 and Windows Server 2008 R2 support in Update 1
- Expanded support for hardware virtualization extensions
From a technical perspective, these features make upgrading to VMware vSphere 4.0 a compelling proposition. So why aren't more companies upgrading? And is the technical aspect the only thing to consider?
Evaluating the upgrade from a business perspective
While VMware can point to new features like the vNetwork Distributed Switch (vDS), VMware FT, host profiles, and third-party multipathing support as technical reasons to upgrade, the truth of the matter is that the need to upgrade is really driven by the business. Do these features offer real business value? The vDS might be cool, but what's the business value? I'd like to look at a few of the new features in VMware vSphere in the light of business value:
vNetwork Distributed Switch
The real business value here is labor reduction, i.e., making it easier to make network configuration changes. In environments with lots of hosts, the time saved by centralizing the configuration of virtual networking can be quite significant. In smaller environments with fewer hosts, the time savings might not be so grand. There's also an indirect cost savings benefit here as well: operational savings resulting from improved consistency. Because the vDS ensures that all hosts have the same networking configuration, troubleshooting is generally easier and the risk of a misconfigured port group is generally smaller.
VMware Fault Tolerance
Do you have workloads that require the uptime that VMware FT can provide? If so, then there is real business value in upgrading to VMware vSphere so that your IT organization can meet the service level agreements (SLAs) that the business needs. Keep in mind, however, that VMware FT in its current incarnation has some noticeable limits that might prevent it from being as useful as it might otherwise be. Specifically, the lack of Virtual symmetric multi-processing (SMP) support means that only VMs with a single virtual CPU can be protected using VMware FT.
Host profiles really shine in larger environments with more ESX/ESXi hosts, but its value in smaller environments with fewer ESX/ESXi hosts might be less dramatic.
Third-party multipathing modules
There is really only one third-party multipathing module available: EMC PowerPath/VE (PP/VE). While PP/VE supports more than just EMC storage arrays, it doesn't support every storage array. The fact that VMware vSphere supports third-party multipathing modules is helpful, but the value to the business is only there if the third-party modules are available and support your storage array.
Expanded resource limits
Sure, VMware vSphere 4.0 can support VMs running up to 255 GB of RAM and 8 virtual CPUs, but do your workloads really need that sort of horsepower? If your business doesn't have the kind of applications that need these expanded resource limits, then there isn't much of a business case for upgrading based on this feature.
While technologists like to discuss "speeds and feeds," understanding -- and being able to articulate -- the business value of new features in VMware vSphere is truly the key to determining whether your organization should upgrade.
Putting it all together
Looking at both the technical and business aspects, I believe that many organizations currently running VI3 should upgrade to VMware vSphere, even if some of the "flashier" features don't necessarily apply. Your company might not be able to use VMware FT, but could your company use a more efficient VM backup process? Improvements made in VMware vSphere "under the covers" enable some third-party solutions, like backups, to become more efficient. I think that VMware's focus on efficiency in this release brings many positives for organizations, but in the end the need to upgrade must be driven by the needs of the business and the value that VMware vSphere will provide.
Scott Lowe is a senior engineer for ePlus Technology, Inc. He has a broad range of experience, specializing in enterprise technologies such as storage area networks, server virtualization, directory services, and interoperability. Previously he was President and CTO of Mercurion Systems Inc., an IT consulting firm, CTO of iO Systems Inc.