Alternative considerations when going virtual

Roger Baskerville: A virtual infrastructure replaces many physical points of failure with one main artery

Shifting from physical to virtual infrastructure is one of the largest changes an IT department may face today. The process harbours no end of pitfalls for the under-prepared.

Methods and tools for migrating an infrastructure from physical to virtual (P2V) depend largely on the size and scope of the project.

One prerequisite is effective consolidation goal-setting, identification of virtualisation candidates and capacity planning.

A virtual infrastructure replaces many physical points of failure with one main artery.

Consolidation of vulnerability can be risky, too. While high-availability strategies may insure against system failure, down-time from planned maintenance and upgrades will still interrupt service.

Ensure mission-critical applications are suitably distributed between servers to avoid clashes in uptime requirements.

Compliance is a notable challenge to be acknowledged in the shift to virtual infrastructure. From the conversion process itself to the continued monitoring of activity for audits and regulatory compliance, information must be tracked accurately and meaningfully.

It is also worth considering the production of evidence that proves detection and limitation of threats such as rogue or misconfigured virtual machines.

A strategic shift to virtualisation may result in a more centralised, secure and easily auditable infrastructure.

A virtual environment obviously has different bottlenecks than its physical equivalent.

During conversion, adjustments are needed to ensure balanced workloads and minimise the possibility of applications competing for resources.

This issue can be confronted and corrected further down the line but it is preferable to start as you mean to go on.

By all means, automate the task as much as possible, but never forget the importance of systems through which you may closely observe the conversion.

This means you can more accurately estimate the completion date of the project as well as form a habit of observation that will serve system maintenance well in the long run.

Ease the inconvenience of the conversion period. In identifying consolidation candidates, have a clear sense of the tolerance for downtime of each server. This is particularly important for migration of production environment servers.

Notable but neglected strategies in this area include creating a separate network for transfers so as not to disrupt operations.

Options such as clearing machines of data held in recycle bins and other unnecessary material will also accelerate the completion of the process while helping prevent virtual machine sprawl.

Speed is crucial in large scale migrations. Large enterprises should carry out as many migrations as possible. Many migrations are carried out on weekends when network activity is at its lowest, so costs, including overtime costs for engineers, are important.

New tools and previous experiences are making each step in the process easier, smoother and faster.

And in time, these migrations will undoubtedly become business-as-usual as the modern workplace evolves.

Roger Baskerville is EMEA vice president at Vizioncore