Leveraging Cloud and Virtualization for Disaster Recovery

1 comment

As the data center continues to become an integral part of any organization, administrators are working hard to find ways to be as resilient as ever. The data center environment is a lot more complex now with many moving parts – all of which are vital to the efficiency foundation of an infrastructure. With cloud computing, more users, and an increase in data – the challenge has become disaster recovery and business continuity. All distributed systems have to be checked and the data points must all be monitored. In working with these more complex data centers, many administrators are turning to the cloud and virtualization to help them create a more robust DR plan.

The reality is this: a well-planned out cloud and virtualization solution can truly help any organization create a more agile environment. There are inherent benefits to working with specific types of cloud models and virtualization platforms. A large part of IT is creativity – that’s why using new types of technologies can help reduce management costs and keep an environment running longer.

The following are some ways that cloud and virtualization can be leveraged for disaster recovery:

  • Cloud for replication. Site-to-site replication has become easier with the utilization of both private and public cloud technologies. With better storage systems and more control over the WAN, organizations are able to better replicate their environments. This can be entire virtual machines, specific databases, or just data points. Furthermore, cloud computing has made disaster recovery much more financially feasible for more organizations. Why? The inherent flexibility of the cloud means you can dictate exactly how much downtime your organization can tolerate and where the costs break even. With that, companies who are trying to stay financial conscious are able to design a solution that fits both – the data center and the budget.
  • Virtualization as a mechanism for backup and recovery. The idea is simple. It’s much easier to recover a virtual machine than it is a physical device. So, administrators with numerous VMs in their environment are working with virtualization-ready backup strategies to keep their data centers agile. Snapshots can be distributed to various points using private networks or cloud technologies. Furthermore, administrators are able to have “mirrored” VMs running at a remote site ensuring an active/active scenario should one of the vital corporate VM workloads fail.
  • Using-software defined technologies. The conversation around software-defined technologies continues to grow where some see this term as an extension to virtualization. Software-defined platforms can range from networking equipment to security appliances. The idea is to create a truly agile environment where various “virtual” networking instances can be deployed. For example, Global Server Load Balancing (GSLB) can greatly assist a DR strategy. By load balancing data centers on a distributed level, administrators are able to seamlessly (and automatically) point users to a different data center should the primary go down. In fact, a similar connection can be created whether that secondary data center is privately held or provisioned by a data center provider. For the end-user, the transition is almost transparent. For the IT administrator and the business manager, this means – very simply – less downtime and faster recovery.
  • IaaS or “Data Center On-Demand.” Cloud and virtualization platforms now allow for the very fast provisioning of needed resources. In a disaster scenario, the ability to recovery VMs and data, quickly, is essential. This is where cloud computing and virtualization can help. Administrators are able to create active/active or active/passive IaaS solutions which can be very cost-effective.  For example, an engineer can set a backup cycle with a given provider. That information is then housed in the form of dormant VMs and backup data. Then, if an emergency happens – the organization can ask the provider to immediately spin up new VMs and attach their data to the newly provisioned environment. Although recovery isn’t immediate, it’s still very quick. Furthermore, organizations are able to adopt a “pay-as-you-go” model so that VMs are only paid for when used. On the other hand, organizations requiring high levels of uptime may simply have the VMs already running. The point behind a data center on-demand platform is the flexibility. Depending on your recovery needs, you can use a provider to help you gain that DR agility.

The current trend clearly points to more information being present in any given infrastructure. This means more users, more devices and much more data. For any organization, there needs to be some type of disaster recovery plan in place. Remember; never place all of your “technology” eggs in one basket. Although cloud computing is resilient, it’s not without its faults.

Over the past year, major cloud vendors have seen serious outages which have caused big companies to become production ineffective for extended periods of time. In these situations, having a plan in place to ensure that users and customers can still access data is vital.

A large part of a good disaster recovery solution is the thought and creative behind it. One of the first steps in creating such a solution is identifying the data and infrastructure components which are required to keep the business running. It’s in these cases that a business impact analysis (BIA) is strongly recommended. Not only will this report show you the most crucial IT parts of your environment, it’ll help you prioritize and plan out a solid DR platform.

Remember, a DR strategy is something you may have to seriously invest in – and hope you never have to use. However, if an emergency even happens, a well-laid out and executed disaster recovery initiative can save time, management overhead and – very importantly – costs associated with experienced extended outages.

About the Author

Bill Kleyman is a veteran, enthusiastic technologist with experience in data center design, management and deployment. His architecture work includes virtualization and cloud deployments as well as business network design and implementation. Currently, Bill works as the National Director of Strategy and Innovation at MTM Technologies, a Stamford, CT based consulting firm.

Add Your Comments

  • (will not be published)

One Comment

  1. AJ Pellis

    Disaster Recovery- and Business Continuity - Virtualization minimizes implementation time for any Business Continuity Plan (BCP) when needed. It can also reduce losses through faster failover and fallback when disaster strikes.