Virtualisation has been touted as the solution to almost all of IT's ills, from resource under-utilisation through to innovative ways to ensure system availability without spending money on expensive hardware, and even quick and reliable ways to implement offsite data recovery procedures.
“Speaking from a continuity perspective, virtualisation, or the ability to dynamically allocate and reallocate resources to applications as required, can play a major role in the IT service continuity world,” says Bradley Janse van Rensburg, solutions design manager at ContinuitySA. “This is especially true when it comes to offsite recovery of critical IT systems. A virtualised server environment will ensure companies have adequate resources on hand when they need it.”
In many IT installations, whether the primary infrastructure or an offsite recovery environment, administrators are known for over-catering the resources installed to ensure they can manage whatever the maximum processing requirements that business expects during peak times. This means that a large portion of the technical resources sit idle for most of the time, 'just in case' they are needed.
In most instances, administrators try to ensure there is more than enough in order to allow for unexpected processing or data storage requirements. This is simply a waste of money and resources. In an IT service recovery situation, the imbalance between capacity and true requirements is even more pronounced as companies rarely test their recovery configuration under load and recovery often has specialised resource requirements (tape restore throughput and database integrity checking as examples).
“In situations like this, virtualisation can add tremendous value as multiple systems can be linked together and addressed as one super system,” explains Janse van Rensburg. “Administrators can assign resources as required, even having the system automatically provision and de-provision them as needed.
“When it comes to recovery sites, since the administrators of virtualised environments inherently have a better understanding of resource utilisation, they have a better understanding of how much processor, memory or disk space an application requires and will find it easier to provision the backup systems when the production installation fails.”
An experienced virtualisation expert will be able to effectively create capacity-on-demand solutions based on the real requirements, including, for example, additional resources for validity checks immediately after the failover. Unfortunately, the ability to move applications and resources around also has a negative impact as it provides an excuse for many administrators to avoid setting up recovery solutions.
“The thinking is that if a failure on one system can be dealt with by moving the applications using it to another fairly quickly, then why bother with a recoverability plan,” says Janse van Rensburg. “For small localised failures, virtualisation does offer some resilience benefits. However, real world networks are at risk to human error and malicious damage, data corruption, cascading system failure and change risk. This is where a separate recovery solution is critical, and within this solution virtualisation can play an important, cost and resource-saving role.”
Share