Contributed By Robbie Wright, Senior Product Marketing Manager, CommVault

IT departments today are looking for ways to get a better grasp on implementing a cloud strategy, and one major motivator encouraging enterprises to invest more resources in the cloud is disaster recovery. However, even though protecting information in the event of a disaster has been a key area of concern for several years, many companies still lack a comprehensive plan and are only budgeting to protect the most mission critical applications. Organizations of every size need the capability to bring an appropriate level of disaster recovery to workloads that have never had that level of protection before, and the barrier is almost always cost.

The public cloud is emerging as the most desired infrastructure choice for disaster recovery. This is true not only for smaller organizations that have been running lots of workloads in the cloud for a while, but also for larger enterprises, who seek to bring cost-effective DR capabilities to more and more workloads.

Virtualization is a key enabling technology in the cost-effective DR equation.  Virtualization begins to remove the requirement for replica hardware - one of the largest cost components of a DR solution - at a secondary site. However, virtualization alone doesn't solve the problem. It is in an organization's best interest to implement modern data protection and management technology alongside virtualization technologies to keep process efficient and costs low. Further, quality virtual machine backup and recovery solutions help bridge different hypervisor or cloud platforms so you're not stuck with a software-defined version of the old physical DR infrastructure problem. Instead, leveraging different, more cost effective platforms for DR and DR test can help. For example, quality data management platforms should enable recovery of a VMware VM into Hyper-V (or vice versa) or perhaps even into AWS, Azure, or other cloud platforms.

To help, outlined below are some tips on how your IT department can leverage virtualization and cloud infrastructure to ensure a disaster recovery plan is set up for success.

Virtualize, virtualize, virtualize

Adjusting the technology first can help ensure a disaster recovery plan starts off on the right foot. Instead of having a mirror image of the hardware at both the production and secondary sites, start virtualizing. Virtualization helps reduce costs associated with the physical hardware (which typically only backs up mission critical apps), and budget can instead be reallocated to bring disaster recovery to applications that didn't have DR capabilities before.

We understand that there are still certain workloads that aren't great candidates for virtualization - at least not in production. However, virtualization is a viable option to run an application long enough to get the physical server back up and running in the event of a failure. A good backup and recovery solution should give your business the ability to recover a workload deployed on physical hardware into your virtual infrastructure on the fly.

Be a copy cat

Once you've virtualized as much as you can, it's time to focus on what's inside and what matters - the data. First, start replicating your backup data from the primary site to the appropriate cloud partner. Second, automate your disaster recovery test process to validate your DR plan. Automation allows you to test more frequently, helping you ensure that your DR process actually works.

One important note - with the cloud, bandwidth for ingress (putting your data into the cloud platform) is usually free, but bandwidth for egress (getting your data back out of the cloud) can be expensive when you're trying to restore large volumes back to the primary site. Instead, consider recovering that workload in the public cloud until you can restore services at the primary site. Then, reconcile the changed blocks back to the primary site when you're ready to failback.

Stay single

When going out on a Friday night, it's good to have a date, but when trying to manage multiple virtualized environments for an effective disaster recovery plan, it's better to stay solo. That doesn't mean you can't play the field with different hypervisors. It just means you should stick to a single data management platform that bridges multiple hypervisors and cloud platforms, and allows you and your IT department to replicate the automated workflow process across multiple types of virtual and cloud infrastructure to improve scalability and eliminate many of the manual steps that used to come with a non-virtualized disaster recovery plan.

Don't forget the camera

While going through all of the above processes, be sure to take out the camera and take snapshots of all the environments. That way, if there are any issues, data can be restored to its original state. Further, with automation and deep integration, this process can help protect the hundreds of virtual machines your IT team may have created, all done within minutes.

Creating a modern virtualized disaster recovery plan can save your business tons of resources and give IT departments a bit more peace of mind knowing that data is safe and easily restorable, with shortened RTPOs AND reduced costs. What other disaster recovery tips do you have to offer?

##

About the Author