According to research firm Aberdeen Group, that is the estimated amount lost for every hour of downtime suffered by the average business when disaster strikes. There is little doubt that a hurricane or flood could bring your data center to a sudden halt, but according to a global study conducted by Forrester Research, over 40 percent of the IT executives surveyed indicated that it was a simple power failure, not a violent storm, which caused their most recent data centre outage.
How likely is it that your enterprise will experience a significant disruption in data centre operations? Forrester reports that during the five-year period beginning in 2003, as many as 27 percent of the businesses surveyed declared that they suffered the effects of a disaster and invoked recovery procedures.
Regardless of the reason for the disruption or the frequency, what matters most is, when disaster strikes, how quickly you can recover your most critical applications and business processes to get your business back up and running.
What’s even more concerning is that according to National Archives and Records Administration in Washington, D.C.; 93 percent of companies that lost their data centre for 10 days or more due to a disaster filed for bankruptcy within one year of the disaster.
Depth of recovery
When a disaster strikes, it is clear that there are serious repercussions to the business. In addition, there could be temporary or even permanent loss of critical data, including customer histories, financial records, marketing databases, email and personnel files. As such, offsite, comprehensive and regular data back-ups are the first step in an IT continuity plan, but in today’s 24 hour business world, protecting against data loss is simply not good enough.
In addition to data loss, disasters may also lead to the permanent loss of physical infrastructure including IT infrastructure. This loss may result in the inability to fulfil existing or new business orders. In most cases, the loss of key IT applications or services can have as negative an impact on the business as the loss of data. To be fully protected, organisations must not only have a plan to quickly restore data, but also the underlying server capacity and the business services those servers support.
Good, better, best recovery
Looking at disaster recovery solutions on a scale of good, better and best, it is certainly possible to say that full and regular data back-up are a good start, but they are far from good enough.
Another option is to use server virtualisation technologies. By encapsulating the OS, drivers, middleware and application software in a Virtual Manager (VM) and decoupling that stack from the underlying hardware, customers can abstract the hardware and create a homogenous pool of servers. This works to ensure each back-up server appears identical to its production counterparts. This in turn allows the VM’s to be replicated to a secondary site and invoked when needed. In this way, virtualisation helps in both maintaining compatibility of hardware as well as in synchronisation to ensure consistency of correct software versions and patch levels between sites.
While disaster recovery (DR) based on server virtualisation can reduce costs and improve consistency, these solutions also create new barriers. DR solutions based on virtualisation, first and foremost, do not support applications running on physical servers. And even today, in the era of virtualisation, large databases and other mission critical applications remain un-virtualised as IT executives are scared about moving key applications to share infrastructure. Yet, paradoxically, these are the very type of applications that require the highest levels of protection. In addition, with disaster recovery using virtualisation, the customer must still maintain dedicated infrastructure, duplicate software licenses and purchase special DR licenses.
The pinnacle of a recovery system is one that handles both physical and virtual servers, maintains configuration consistency at both ends of the wire and automates the recovery process. As a result, enterprises can reduce their recovery time objectives account for application priorities and dependencies and guarantee the recovery of an exact copy of the environment. This includes policies and configurations for the server, OS, applications, network and storage connections, which all ensure the core IT services have the capacity, availability and response time to meet service levels required by the business.
Impact of complexity on recovery
Yet, as IT has become more important to the business, it has also become more complex, and recovering critical IT services has become an even more difficult task. Today’s IT managers have to orchestrate across virtual, physical as well as cloud environments to ensure all the critical services to the business are restored. For most organisations this is a complicated, manual and time-consuming task in which success is far from guaranteed.
This was evidenced recently when, in October, RIM’s BlackBerry service went down after a core switch failed in its email and web services infrastructure. The outage or “delay” as RIM described it, affected users throughout the world as the backlog of data clogged RIM’s systems. According to the company, it was designed to failover to a back-up switch, but the failover system "did not function as previously tested." In total, RIM’s service was down for about two days, yet in its most recent earnings report, RIM announced that it will be taking over half a billion dollars in charges related to the failures in its services this fiscal year.
The point is simply that disasters have far reaching impacts. Yet if businesses are going to confidently meet customer’s expectations, recovery practices must change. Today’s DR best practices must account for the growing complexity of IT, they must assume a diversity of platforms and, most importantly, they must provide guaranteed, fast and verifiable recovery through digital automation of the recovery run book.
Yet despite the industry’s best efforts, IT strategists still encounter several obstacles on the path to creating a cost-effective and reliable disaster recovery plan. These obstacles include siloed processes, prohibitive capital costs, the inability to test with frequency and ineffective recovery solutions that actually increase operational complexity and add new obstacles to a recovery process that meets today’s business needs.
We see these frustrations and desires reflected everyday with our customers. These organisations understand that for their customers, there is no such thing as acceptable downtime. Revenue, reputations and brand all hinge on the ability to provide always-on services to the customer and employee base. To make this possible these organisations have embraced the concept and creation of a “data centre fabric”. These fabrics abstract network switches, servers, storage connectivity and even database and virtualisation software and provide a digital copy of the data centre. That digital copy not only contains the configuration information of the network and server infrastructure, but also contains address, policy and connection information on the distributed environment. This file can be thought of as a recording of the data centre and, if a disaster does occur, this recording can then be run at the recovery site to automatically recreate an exact replica of the primary site – quickly and every time.
It is only through such holistic and automated approach to recovery that organisations can be certain they can reduce the risk and impact of infrastructure failures. And as enterprise IT departments around the globe face growing demands for more reliable and resilient services, this simple concept is changing how IT managers deploy applications, manage data centre capacity and ensure core IT services can be restored in minutes.
 ShareThis
Tags: BC/DR |
