Choosing and Managing Your Hypervisor Solution

A Contributed Article by Sanjay Castelino, VP and Market Leader, SolarWinds

When preparing to deploy and manage multiple hypervisors, there are several things to take into consideration. Functionality and cost may be the first elements that come to mind, but skills and cloud strategy also come into the mix when deciding on a hypervisor solution.

Cost and Functionality:

Many organizations use a variety of hypervisors to manage their virtual environments. Cost and functionality are typically cited as reasons for adopting different systems. While VMware vSphere has long been the "gold standard" for hypervisors, IT departments with limited resources have been able to diversify their hypervisor purchases since not all areas require the bells and whistles vSphere provides. For non-critical parts of IT such as development, test and QA, many organizations turn to lower cost alternatives like Microsoft Hyper-V, where functionality has been steadily improving, and is in fact expected to significantly close the functionality gap with vSphere with its next release (Hyper-V 3). In a recent SolarWinds survey of 406 IT professionals, 55% reported that they already have or have plans to deploy Hyper-V in production or non-production environments, and 43% said that they believe Microsoft will bridge the functionality gap between Hyper-V and vSphere.

To lock in, or not to lock in?

Using more than one hypervisor does mean a greater amount of complexity, but enables the freedom from vendor "lock-in" - what some IT professionals consider to be an equally risky situation. Vendors are looking to align companies to a complete stack, but customers may not be comfortable with their infrastructure management vendor also being their systems management vendor. They may be happy using vSphere, but may not want to pay for additional "VRAM," and particularly don't want VMware making recommendations for how much VRAM they need to run their systems. For many customers, it's not just a question of which hypervisor you choose, it's what you're going to be doing with it. Is disaster recovery part of your plan? Private cloud and self-service deployment? What are the things the hypervisor is ultimately enabling? Vendors are now grouping the hypervisor, disaster recovery, backup software, private cloud enablement software, et cetera, to create whole suites that can lock you into their stack.

Skills:

Many virtualization administrator roles have grown out of the traditional sysadmin role, meaning that they have backgrounds typically focused on server management, and likely know Windows very well. That Windows training will serve them in good stead as Hyper-V continues to make its way into the enterprise, though even vSphere is Windows user friendly. The flattening of the technology stack will continue regardless of which hypervisor you use, and will continue as more organizations embrace private cloud. This means that regardless of choice of hypervisor, virtual administrators need to have a broad set of skills across server, storage, network and possibly applications as the hypervisor touches all of these areas and continues to integrate with them ever more deeply.

Cloud:

It is also important to consider your organization's future cloud strategy. Standardizing on one hypervisor may be appealing as it will allow you to reduce the number of moving parts, making it easier to move to that particular vendor's cloud. While Hyper-V is compatible with Microsoft Azure and VMware is compatible with the VMware Cloud options, using one exclusively may limit your options for future public cloud adoption. Hypervisor and cloud interoperability standards are still in their infancy, and the reality is that settling on one vendor for your hypervisor and private cloud could potentially limit your choices down the road. Sure, you can convert a virtual machine from one format to another, but remember that the complexity of moving arises in all of the services and applications that are built around the hypervisor - and not just the particular run time format.

Management:

Part of the challenge of managing multiple hypervisors is that each comes with its own proprietary software, vocabulary and features. Users want to be able to manage both from a single installation or interface rather than swiveling between them. While separate software may not be an issue as long as it's being used in separate departments, problems can arise, for example, when something is run in development and now needs to be run in production. Unless those two areas are under the same hypervisor (and as referenced before, they may not be), it is difficult to compare the before and after from development to production.

Users need a vendor neutral solution that provides a single pane of glass view into multiple hypervisors, rather than relying on vendor-provided tools to deliver the capabilities they need. Single pane of glass virtualization managers give users the ability to see where their hypervisors are similar and where they are different. Where they're different, a virtualization manager allows you to see either the V-Sphere or Hyper-V functions independently. It is important for IT to be able to plan to capacity and see how much resources an application is using in development say on Hyper-V, and what that translates to in production on VMware vSphere, for example. As the stack continues to flatten, a more comprehensive view of the organization's virtual environment will be crucial to keeping things running smoothly.

Storage is another crucial element of management for any hypervisor. Whilst VMware and Microsoft have varying levels of instrumentation built in to the hypervisor to measure storage performance, it is critical in either case to be able to tie what is happening at the virtualization layer to what is happening in the physical storage layer underneath.

Whether you elect to pursue a single or multiple hypervisor solution for your organization, evaluating how the virtualization fits into your overall IT strategy and what needs it must fulfill will help you make the best decision for your business.

###

About the Author

Sanjay Castelino is a VP and Market Leader at SolarWinds, an IT management software provider based in Austin, Texas. Sanjay leads the company's initiatives around its end-to-end IT solutions for network, SIEM, storage and virtualization management. He is responsible for our product strategy and go-to-market efforts in these markets.

Previously, Sanjay worked at NetStreams where he held the position of VP of marketing and business development where he oversaw all marketing, product management and OEM business and strategic partnerships. He was also the VP of product marketing and management at Motive where he helped to lead the telecom business through its inception and acquisition of Broadjump.