Virtualization and Cloud executives share their predictions for 2013.  Read them in this VMblog.com series exclusive.

Contributed article by Shannon Snowden, Sr. Technical Marketing Architect, Zerto

2013 is the Year of 'Disaster Recovery to the Cloud'

Disaster Recovery (DR) to the Cloud will be the top reason Enterprises will adopt a public or hybrid Cloud solution in 2013.

‘Something Something as a cloud Service' (SSaaS) has been a buzz phrase for a few years. But the reality is transforming applications to be cloud hosted is a long-term undertaking that is competing with existing solutions that work well enough today in the Enterprise. I don't think something is going to pop up that will make Enterprises jump into the cloud in 2013. There may be pockets of adoption, but nothing wide-spread from an Enterprise perspective - with one exception.

The one exception to that prediction is DR as a Service (DRaaS). What I have seen that actually works very well today is DR to the Cloud.  The technology is available and proven to be very reliable.

Now I believe it's a matter of awareness building to a wider audience that production-ready Cloud providers exist that are successfully operating as both sites, a second site or hybrid site for Enterprises in sectors such as communications, financial, healthcare and manufacturing.

Learning from history can help predict the future

History shows us that organizations wanted virtualization for all the cost savings and portability benefits, but it needed time to become a proven reliable solution. It certainly did prove to be very reliable and now the most business-critical applications are virtualized.

The obvious next step would be to build on the hardware abstraction and portability that virtualization offers and leverage it for disaster recovery to a remote site or multiple sites. There also certainly is a desire from many organizations to use a cloud provider for their DR site because of the signification cost savings of not needing a second datacenter. However, this hasn't really had a widespread rate of adoption yet. 

A brief analysis shows that something was wrong with the solution being offered.

Components Needed for DR to the Cloud Success

Let's review what service providers must have in place to offer an attractive DR to the Cloud solution.

1.       Hardware agnostic: hypervisor-based replication can replicate from any storage to any storage between sites, increasing efficiencies and lowering the TCO.

2.       Secure: at the production site as well as the replication site. Today, BC/ DR solutions include features throughout the information chain to secure them - meeting high standards for enterprise-class, mission critical applications. User authentication, communication to the virtualization appliance, networking and port configuration, network encryption and stronger roles and permissions granting are all in place in today's clouds, making them more secure targets for application replication than many private enterprise datacenters.

3.       Multi-tenancy: Complex management issues are introduced when replicating multiple tenants to a single cloud. When replicating from separate company sites to one Cloud Provider site, redundant hardware and management software is required at the replication target. In many cases, this redundancy takes away the flexibility and scalability the cloud offers. 

4.       Multi-site capable: CSPs have customers in different locations and multiple data centers serving them. If the solution cannot support bi-directional replication across multiple sites, the CSP cannot offer the service. Today, Cloud DR solution providers offer multi-directional replication across as many sites as the infrastructure requires.

5.       Integration with cloud management platforms: Many CSPs deploy customized cloud management platforms and require BC/DR to be managed, provisioned and automated through these platforms. This requires tedious integration support between the two. Cloud Service Providers can now report and act on all of their customer environments from a central console, including initiating automated failovers and failover testing. There is no need for manual correlation of activities across the environment.

6.       Aggressive recovery Service Level Agreements (SLA): Replication solutions need to be able to support mission-critical applications with Recovery Time Objectives (RTO) of minutes and Recovery Point Objectives (RPOs) of seconds, as well as other applications with less aggressive recovery needs. Whether a few VMs or thousands, new solutions can scale to effectively support mission critical applications, and have extremely short RTOs and RPOs.

Historically, cloud providers were missing several of these pieces, so the adoption rate of any DR as a Service was very low.

If you build it, they will come

But that has changed. Over the past year, we've had the pleasure of working with many cloud providers that have adopted the technologies and processes to meet the above requirements and we have enabled many successful production deployments with Enterprise customers being protected.

We have so many production stories of businesses getting real value out of DR to the Cloud that are similar to this example below.

During Hurricane Sandy, Zerto customer Affigent LLC, a Herndon, VA-based technology consulting firm, used a cloud DR solution provided by Integrity Virtual IT (an IT outsourcing company in Silver Spring, MD) to protect its business critical CRM and ERP applications. When they were certain the hurricane was in the path of their primary datacenter, they failed over their VMs to a second datacenter that was out of the storm's path. Affigent continued to fully operate without any data loss in the secondary datacenter until the storm passed.  They migrated back to the primary datacenter when they were certain the primary datacenter was ready. Affigent was able to conduct business during the storm without interruption, utilizing their tier-one applications running in Integrity Virtual IT's cloud recovery site.

We've seen our customers and partners become very comfortable with using the cloud for DR and seeing their tier-one applications running in the cloud for recovery purposes. In time, these companies already comfortable running applications in the cloud for DR purposes, will permanently migrate more and more operations to cloud environments.

These customers now know that moving or failing over their mission-critical VMs and applications in the cloud is not a futuristic prediction predicated on something that needs developed in order to exist.  They've already done it, and they're quite likely to do so again. 

It reminds me of several years ago when virtualization finally hit the tipping point from being deployed in test labs only as a proof of concept to being deployed as a trusted solution into production.

DR to the Cloud is at that tipping point. As more Enterprises consider and actually try DR to the Cloud, I think it will become the top reason they decide to leverage cloud services.

##