A Post from My Colleague John Morency: What Can be Learned about Cloud–based Recovery in the Aftermath of Hurricane Sandy #Sandy
What Can be Learned about Cloud –based Recovery in the Aftermath of Hurricane Sandy ???
John P. Morency, Research VP
Given that the widespread availability of public cloud-based recovery services is only a few years old at this point, the opportunities to demonstrate their utility following the occurrence of major disaster events had been somewhat limited.
Hurricane Irene (a late 2011 event) and Hurricane Sandy (which occurred this past week) certainly have changed that perspective, but only to a certain extent. The success stories of organizations that failed some or all of their IT operations over to their provider’s cloud either prior to or during Hurricane Sandy have already begun to appear in the trade press, blogosphere and various social media outlets such as Twitter. Similar anecdotes were published in the aftermath of Hurricane Irene last year. However, while these success stories have undoubtedly been instrumental in helping make the case for cloud-based recovery, by no means do they constitute conclusive evidence for all businesses.
First, it’s important to realize that a majority of existing recovery-as-a-service production instances support a :
a) bounded number of in-scope VMs (primarily VMware-based);
b) very limited number of production applications (typically no more than six);
c) bounded amount of production storage (less than five terabytes)
This means that the elapsed time required to fail over production operations from the premise to the cloud is far easier to meet than for configurations whose scope includes hundreds of VMs, applications and/or terabytes.
Secondly, Gartner earlier this year found (through provider surveys) that a number of RaaS providers thus far have had limited experience (greater than one and less than ten customers) with successfully orchestrating IT operations failover following the occurrence of a disruptive event as well as failing back to the premise following the successful conclusion of recovery operations. This will obviously improve over time, but prospects should not consider this experience as a slam dunk for all providers today.
Thirdly, it’s also important to point out that, apart from the large providers, the number of provider data centers that support a given RaaS offering may be somewhat limited (i.e. no more than two or three). In addition, the geographical distance between these facilities may also be limited (e.g. less than a few hundred miles). Given that Hurricane Sandy impacted nearly a third of the continental US, this limited distance may not be sufficient to 100% ensure that all cloud providers can maintain operations during similar events (and there definitely will be similar events in the not too distant future).
Lastly, it’s always important for potential customers to verify not only distance between provider data centers, but to also always ensure that candidate providers provide concrete evidence of the extent to which their facility operations have been sustained both during and following a major disaster event.
That said, I agree that RaaS has a great future and that providers have made enormous strides over the past few years in both growing their customer bases and the resilience of their internal operations. However, it’s also important to not conclude as a result of post-hurricane coverage that the RaaS industry now has the breadth and depth of experience that make RaaS a compelling choice against other viable alternatives. The industry may well reach that point in a few years’ time, but that is definitely not the case today.
Category: advisory bcm-process event technology
Tags: business-continuity-management business-continuity-planning business-resiliency cloud-computing cloud-based-recovery disaster-recovery hurricane-sandy it-disaster-recovery john-morency raas recovery-as-a-service recovery-in-the-cloud sandy virtual-machines virtualization vm
Comments or opinions expressed on this blog are those of the individual contributors only, and do not necessarily represent the views of Gartner, Inc. or its management. Readers may copy and redistribute blog postings on other blogs, or otherwise for private, non-commercial or journalistic purposes, with attribution to Gartner. This content may not be used for any other purposes in any other formats or media. The content on this blog is provided on an "as-is" basis. Gartner shall not be liable for any damages whatsoever arising out of the content or use of this blog.