Server virtualisation comes with some very nifty tools. Microsoft Virtual Server has host clustering. XenServer...
has XenMotion built in, which allows you to move a virtual machine from one XenServer to another without downtime. VMware Infrastructure 3 (VI3) Enterprise Edition includes VMware HA (High Availability), VMware DRS (Distributed Resource Scheduling) and VMotion. Now there's a collection of tools that spell 'uptime.'
Let's take a closer look at the tools in VMware VI3. VMotion gives you the ability to seamlessly migrate a virtual machine off one host and onto another without disrupting the system (or application) availability to the end users. VMware HA will allow an administrator to have their virtual machine restart in the event of a failure. VMware DRS will automatically have the virtual machine start on another host with sufficient computing resources to support its function. In additon, VMware recently expanded VI3 to include Storage VMotion, which will permit the migration of the physical storage layer of a virtual machine between LUNs or arrays.
But exactly how does this assist my backups?
Simply put, a backup is a point in time that you back up your data or system for future recovery. Historically, backups have been used in the event of a system failure, application failure or a disaster. However, backups have evolved so much over the past years that they now form an integral part of the "business as usual" function.
Let's go back to what a backup is used for: system recovery.
In the enterprise arena where active/active data centres are not uncommon with high-speed communication links, the traditional backup function is moving down the priority order for IT organisations.
With the server virtualisation tools discussed above, a virtual machine can be run anywhere, anytime, all of the time. And where this technology complements other vendor systems, such as storage array based replication (for example, the EMC SRDF family, EMC MirrorView, HP Continuous Access and NetApp SnapMirror), which are exceptionally efficient in getting "ones and zeros" between distributed sites, you can soon see that backing up the system becomes a second string to your bow rather than your only string.
What about the SMEs of the world that do not have the luxury of multiple fibre links between globally diverse data centres?
Different challenges are being met through virtualisation. Turning a physical server into a virtual machine, which in simple terms makes it a file on a VMware File System (VMFS) volume on common or shared storage, is something we can all get our heads around.
How do we back up files? What is the quickest way to back up files? The easiest way is with a snapshot of the volume or LUN in the array or by an appliance that replicates between DAS servers. Now you only have to meet the daily recovery point objective (RPO) for disaster recovery or business continuity planning.
About the author: Andrew McCreath is an Engagement Partner with Glasshouse Technologies (UK), a global provider of IT infrastructure services. He has more than 16 years experience with infrastructure and management information systems. Prior to joining GlassHouse, Andrew managed multi-million-dollar projects while employed with Accenture, Credit Suisse First Boston, Kimberly-Clark, Société Générale and EMC, and is currently specialised in server virtualisation and data centre consolidation.