Close

Seeing Through the (DR) Clouds

Brian-118By Brian Reagan, VP Product Strategy & Business Development  – For a time, I worked for a company who provided disaster recovery services to some of the largest companies in the world. A significant percentage of these were using tape-based means of recalling data from underground bunkers and recovering servers and applications via a scheduled appointment in a shared data center. We referred to this as “traditional DR”, which was really a polite way to avoid discussing more efficient and effective means of transporting, restoring, testing, and using data.

There remains a multi-billion dollar market for DR services, but incumbents are increasingly under assault by new thinking that goes something like this: “why can’t I just electronically copy my data from my data center to the cloud, so if something goes wrong locally, I can restore from the cloud-based copy?” It’s a valid thought, particularly given the cloud’s elastic ability to spin up compute and storage resources on-demand. And, its what has led many of these incumbents to increase their investments in cloud.

DR in the cloudA few years ago, AT&T Labs sponsored research with University of Massachusetts Amherst* to evaluate, among other things, the economic benefits of disaster recovery as a cloud service. In this study, they compared the costs for replicating both a multi-tier web application and a more robust data warehouse to both a Public Cloud as well as a private co-location site. The cost savings in the cloud for the multi-tier web application were substantial, nearly 85% less than the comparable co-lo solution. This was largely due to the ability to leverage on-demand, pay-as-you go resources in the cloud versus the need to maintain physical hardware in the private facility. There were also cost savings in the cloud for the data warehouse use case, though the difference was much smaller, primarily due to the requirement for more persistent data storage and compute in the cloud.

DR as a service is finally ready to cross the chasm, particularly as a crop of new capabilities like Actifio Resiliency Director™ come to market. The ‘holy grail’ of push-button automated, rapidly executed, assured tests and actual recoveries of large-scale (hundreds of VMs) is here. Traditional DR is dead, long live assured recovery.

* Disaster Recovery as a Cloud Service: Economic Benefits & Deployment Challenges. Timothy Wood, Emmanuel Cecchet, K.K. Ramakrishnan, Prashant Shenoy, Jacobus van der Merwe, and Arun Venkataramani.