A New Look At The Data Center “Three Rs” Of Business Continuity


Written by jppersson on November 6, 2008 in Business Continuity, Disaster Recovery.

As is true in most areas of IT some things are new and some things remain the same. This is certainly true of disaster recovery/business continuity. We simply cannot move ahead with new recovery strategies if any lingering problems remain. What this post suggests is that there is a logical progression when dealing with data center backup and recovery. The key is to make sure that the proper steps are followed.

This brings to the discussion the  three Rs of data center disaster recovery.

  • R #1: Reliability – It is always interesting to move ahead with new technology when it becomes available. New SAN equipment, virtual servers, etc. are definitely attractive. But R1 must ensure that before any new technology is implemented the data center does indeed have a stable and very reliable environment. Downtime and system outages along with instances of corrupt database files is a sign that moving ahead should be postponed. Focus on a very reliable and stable data center environment first. Otherwise, the results of new technology are compromised by inaccurate and/or incomplete data. For example, it is not all that infrequent that an audit finds new servers are not being backed up.
  • R #2: Replication – The most widely used method of recovering systems after a failure is to use backup tapes. They should be stored off-site. Unfortunately, it is not uncommon to find that daily tapes are not sent off-site daily. This opens the recovery process up to the possibility of a significant amount of lost data. This is definitely not good! In other instances, tapes simply can’t be read. This can largely be solved by implementing a system of critical data replication. Usually, but not always, this is done to another site. While this can still be expensive, the costs are coming down especially for network capacity. In fact, some solutions are using the internet cloud as a transport method. It is a good idea for data center recovery plans to look into replication. Even on a small basis, it is worth the effort. Think about the database systems that require multiple terabytes of recovery and you quickly see how replication is faster than tape.
  • R #3: Resiliency – This has become the current buzzword within the BCP arena. The definitions abound but generally it means the ability to return to a prior state (or prior form) quickly. This capability takes into account the first and second of the 3 Rs. The ways to achieve resiliency include redundant recovery centers, backup networks, clearly defined critical business processes, and semi-transparent method to switch operations to the backup facility. This is a tall order but one that should be in the planning process for all data centers that support business or mission critical operations. That most likely includes all data centers.

When maintaining, testing and planning BCP upgrades for the data center recovery strategy it is a good idea to periodically touch base with the 3 Rs.

======================================

Jan Persson is the author of the GO.RECOVER-Data Center Disaster Recovery Template – a powerful yet easy-to-use tool for under $100.

Be Sociable, Share!

Tags: , ,