By Allen Mitchell, Senior Technical Account Manager, Mena at CommVault Systems
Beyond the obvious cost savings due to consolidation and reduced operational overheads, businesses are now realising that they can improve the availability, reliability and even performance of their mission critical applications by running them in a virtual context.
Gartner recently forecast an expectation for x86 server virtualisation to double from 40% in 2011 to 80% by 2016. While the uptake of virtualisation is encouraging, data protection and management in the virtual environment continues to be a challenge for most organisations. Most existing tools for backup and data protection of virtual machines are limited either in terms of scale, recovery capabilities, scope and support or application integration.
Simply force fitting legacy solutions, which are not optimised for the new virtual environments, only adds to cost and increases the potential risks to overall business.
What CIOs should know about virtual server data protection
One of the biggest misconceptions in IT is the equation between “Best of Breed” and “Point Solution”. Many vendors now offer data protection solutions which they claim are “purpose built” or “optimised” for the virtual world. Typically, point solutions offer data protection for only a subset of the overall data centre – in this case the virtual environment- which means these solutions lack support for what are potentially key elements of a comprehensive data protection strategy.
Consequently, multiple solutions are required to ensure adequate data protection for the entire data centre. The drive to virtual infrastructure is about reducing cost by consolidating resources, maximising utilisation and centralising management. It is thus counter-productive to completely contradict this strategy by adding multiple, complex solutions for data protection and management.
CIOs should be aware that many purpose-built solutions run into real trouble when users attempt to scale beyond fifty or a hundred virtual machines, significantly impacting performance of production hosts and applications. Choosing a solution that doesn’t understand the scale and performance requirements of the converged virtual infrastructure can result in unnecessarily expensive, over-designed solutions or in critical application and virtual machine data that are left unprotected.
Finally, given that by the end of 2012 an estimated 58% of all x86 server workloads were virtualised, it can be deduced that many organisations have already virtualised the basic areas of their infrastructures. Therefore, for most organisations, achieving the goal of a fully or 100% virtualised environment means targeting Tier 1 applications. This puts data management squarely in the critical path of the virtualisation initiative.
IT leaders should be asking tough questions designed to ensure they have a data management strategy that comprehends the drive to a fully scaled out virtual infrastructure, in order to avoid the need for a costly, time-consuming data management redesign. Here are the top three:
Does the solution align to the long term goals of the business?
The virtual platform and initiative is strategic. The supporting cast of infrastructure components and tools- including data protection – needs to align with this strategic direction. If the goal is to become 100% virtualised the data protection solution must be ready to handle the scale, application integration, recovery, and access requirements to make this happen.
Does the solution help reduce cost?
Let’s face it, the primary objective of virtual platforms is to do more with less and ultimately reduce costs across the board. The data protection solution for the organisation should align with that compelling vision. A data protection solution that spans the entire data centre – physical as well as virtual environments – should help ease the transition from physical to virtual.
Additionally, a data protection solution that is fully integrated from the ground up to deliver all the requirements of the virtual platform and does not require additional bolt-on components, 3rd party add-ons, or costly and complex scripting integration projects will help dramatically reduce the overall costs for protecting and managing data across the organisation. CIOs should be looking to choose a solution that not only helps facilitate, but actually accelerates that transition by also delivering scale, scope and recovery in the virtual platform.
Does the solution help reduce business risk?
Beyond reducing costs, we’re now seeing that business can actually improve performance and flexibility and reduce business risk by virtualising more workloads and centralising management and operations. The same is true for data protection and data management.
A centralised solution that spans the data centre and encompasses all data protection and management requirements will eliminate the need for multiple point solutions that must be synchronised with complex scripts and manual intervention – ultimately resulting in more reliable data management processes that ensure information is where you need it, when you need it, and in this way, reduce risk to the business.
A poorly architected data protection solution can stop a virtualisation deployment dead in its tracks. IT leadership should be looking to storage and data management teams to ensure they are ahead of the challenges created by the tidal wave of consolidation and infrastructure convergence that is upon them in the form of server virtualisation.
As organisations look to build out large private and hybrid cloud environments, the data protection and data management strategy must be ready for that level of scale and efficiency in order to avoid the costly re-architecture which many IT leaders are now realising is needed.