While business continuity in the IT industry has been around for a while, it is still not widely understood, accepted or implemented. There have been many abject lessons about why every business should have a business continuity plan from natural disasters to terrorist activities to something more mundane such as a disgruntled employee.
IT business recovery in the modern data center can take many forms. Legacy implementations include tape-based recovery, off-site storage and copies of critical data to ensure data recoverability within defined timeframes. The aim of IT business continuity is to ensure that IT resources required to keep a business running are available within defined tolerance limits. It is not to be confused with high availability that focuses on redundancy of hardware components. High availability is a component of business continuity that enables hardware to be more resilient.
I’ve been in the industry long enough to see the dominance of tape-based recovery as the only viable technology to instantaneous failover of IT resources to another data center entirely. Customers who exercise their business continuity plans on a regular basis often come to the realization that they have reached another plateau in their recoverability. A well maintained business continuity plan includes exercising all or part of the plan.
One of our customers had exercised his disaster recovery plan and the results clearly showed that his current technology was inadequate to the changing needs of his IT continuity plan. The ensuing technology evaluation clearly showed that the newest storage technology coupled with array-based replication technologies significantly decreased the time to recover. The old system took over 48 hours to get the data to the offsite data center and begin the recovery of the data. The recovery was broken down into these main steps:
- Retrieve data storage media from offsite storage and ship media to recovery site (1 day)
- Rebuild the catalogue images (1 day)
- Start restoring the systems detailed in the disaster recovery plan (½ day)
- Restore data sets (1 to 5 days)
The new solution used replication technologies and not only had data sets that were more current, but was able to make the data available within hours of detecting a failure. The steps to recover using this technology are:
- Break the replication pairing and activate the cloned file system (20 minutes)
- Mount the data (10 minutes)
Let’s take a closer look at what IT business continuity really means. An IT business continuity definition includes timeframes, infrastructure and methodologies. These details dictate not only the what, but the when and the how IT infrastructure is to be recovered. As technology changes over the lifecycle of IT infrastructure, new continuity and recovery paradigms can be put in place.
Replication is a cornerstone technology for modern data center recoverability and IT business continuity. Coupling this technology with virtual infrastructure technologies enables recoverability on a scale never thought possible without significant investment in redundant hardware and networking technologies.
Many businesses today utilize virtualized data centers where the majority of the IT resources are virtual. Workloads are shared amongst virtual systems, virtual networks and virtualized storage. This proliferation of virtual infrastructure has enabled new paradigms of business continuity, leveraging the latest technologies to enabling aggressive recovery definitions that were once reserved for enterprise consumers with business critical applications. These enterprise applications were so important to large corporations that the loss of these services for even a few minutes could potentially be a loss of revenue measured in the millions of dollars. Virtual technologies today can perform individual system, groups of systems, and entire data center failovers to an alternate location within minutes.
While the technologies have matured, the business requirements and acceptance of IT business continuity is sadly behind what can be done. Many businesses consider that as long as they have “something,” they’re OK. It’s not until something actually happens to their infrastructure that they give business continuity and all that it entails real consideration. I was fortunate to work for a few customers who had a mandate from the executive committee to have viable, living business continuity plans. These enterprise customers were committed to ensuring the continuity of their businesses. I have also been in environments where “it was on the books” for later that year. I sincerely hope they don’t have a catastrophic failure before they get to it.