Despite the broad adoption of server virtualisation, many firms still struggle with the technical and organisational changes that virtualisation brings. We see IT shops facing three major challenges: getting business and executive buy-in, motivating application owners to move to virtual machines, and restructuring the IT team to achieve maximum efficiency in a virtualised environment.
The best practices your organisation must follow in order to reap the maximum efficiencies and cost-savings of virtualisation adoption include:
BEST PRACTICE NO. 1: GET THE BUSINESS ON YOUR SIDE
Many firms struggle to convince application owners that their applications will run at the same level (or faster) on virtual machines as they did on physical machines. Another complication: virtualisation is usually a bottom-up initiative, so executive support often times does not come easily. In order to quell fears and attain executive and business sponsorship, IT professionals can follow a few guidelines.
It's important to prove that virtualisation works. Start small with a trial of the development or test environment on one of the company's non-critical apps. Before beginning, set up achievable service-level agreements (SLAs) and then prove that the virtual machines can reliably meet or outperform them.
It may be beneficial to bring in a consultancy to build or vet your business case before presenting it to the business. It helps to have a second opinion, as you may overlook cost savings beyond the server hardware. Plus, you'll appreciate using a validated methodology when defending harder-to-quantify cost savings such as improved disaster recovery capabilities.
Use chargeback as an incentive to the business side to go virtual. For instance, offering a managed virtual machine at lower cost than a managed physical machine may be all the incentive you need to drive application owners towards virtualisation. If you don't have a chargeback process already set up, consider "memoback", where managers receive notifications on what they're consuming, either in dollar amounts or in lists of resources.
And, remember to offer better system availability as a reason to go virtual. virtualisation offers a range of availability and disaster recovery capabilities that can provide quick restoration or the ability to failover to an alternate site.
BEST PRACTICE NO. 2: GIVE YOUR STAFF THE TOOLS THEY NEED TO BE SUCCESSFUL With new technologies come new required skills, you must decide how to get your IT staff up to speed. Some firms hire a new staff member as the resident virtualisation expert, others hire a consultancy to drive the process; but, most find ways to complete the project with their own internal staff.
No matter which route you take, it's helpful to use a combination of vendor and peer-to-peer on-the-job-training. Most vendors offer combinations of on-site or Web-based training. If not, you can minimize costs by sending only a few members of the team to training with the vendor. Then, those who were trained by the vendor can train their peers on the job. However, if it's a particularly difficult migration, it may be in your best interest to bring in a contractor or consultant for the specific project.
Keep in mind that staff members need to be trained not only on new technical skills, but also on some of the softer skills they will need when running the virtual environment, such as communications, planning, and negotiating. These skills are especially important when managing virtual resources that are shared among multiple applications and ensuring that resources are never overcommitted and SLAs are established and met. If you're running a large enough virtual environment, you may want to consider hiring a capacity manager.
And, when bringing in someone new, consider hiring those with relevant experience over certified professionals. A good virtual infrastructure administrator needs broad experience with server, storage, and networking technologies, as well as experience with forecasting and capacity planning-many of these competencies can only be acquired with years of hands-on experience.
BEST PRACTICE NO. 3: IMPROVE AGILITY BY MAXIMIsING PRODUCTIVITY
Most firms don't have trouble reallising the hard dollar savings associated with virtualisation, but you can use several virtualisation strategies to make your team more productive and in return increase operational efficiency.
Grow your environment without increasing your staff. Logically, this makes sense. When you're able to turn off a box because its workloads have been consolidated or virtualised, you would not throw it away. Instead, you are likely to use that box to expand your environment-and the same should be done with employees. We see firms growing the capacity of their infrastructure environments by 25 to 60 percent with the same number of staff.
Faster server provisioning and a faster server request workflow both promise huge efficiency gains, but only if you re-architect the process around virtualisation. So you should optimize your IT management processes for a virtual environment. This can reduce provisioning time from several days (or weeks) to a few hours, without even including procurement time. You will also need to rethink processes such as backups and software distribution if you want to reap the full benefits of an efficient virtual environment.
Finally, standardise on a hardware configuration or at least a single vendor. Standardizing on a single vendor will save time and money, as your staff will only have to coordinate with one vendor and this facilitates the reallocation and reuse of servers. Because virtual servers are very portable across hardware platforms, the problems historically associated with a single vendor are greatly reduced.
Use virtualisation as a driver toward more centralized IT services. Because virtualisation requires a centralized environment to reach the economies of scale required for maximum cost savings, organisations can leverage this to drive towards more centralized IT services. Start with a shared test/development environment to show the cost savings that can be achieved when the disparate IT departments combine resources instead of implementing virtualisation in many smaller environments.
Galen Schreck is a Principal Analyst and Rachel Dines is a Senior Research Associate at Forrester Research.
Join the CIO New Zealand group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.