Getty Images/iStockphoto


Maximize the benefits of virtual disaster recovery

Compared to physical setups, virtualization can accelerate and streamline DR processes. Automation, however, is essential to achieve those benefits.

The primary goal of disaster recovery is to get a viable service up and running as quickly as possible. This minimizes disruption to key business processes and ensures a company can continue to generate revenue.

Virtual disaster recovery -- a DR model that enables failover to virtualized workloads -- helps IT teams meet these goals. IT admins can use replication to make a real-time copy of virtual machines in a secondary location. Compared to traditional DR methods that rely on physical resources, virtual DR offers greater speed, efficiency and flexibility.

That said, to benefit from virtual DR setups -- and to ensure they work as intended -- IT teams must invest in the right tool sets and embrace automation.

Benefits of virtual disaster recovery

Every task IT teams must perform to bring DR systems online increases the time to recovery -- as well as potential revenue loss. The time and speed with which admins can spin up virtual infrastructure lends itself well to almost instant DR. As long as there is capacity at the DR site, it is relatively quick and easy to stand up DR resources.

As part of that instant-on virtual capability, failover becomes more streamlined, less resource-intensive and less error-prone. But most importantly, it occurs much more quickly than it can with traditional DR -- potentially with sub-five-minute recovery time objectives.

Another major benefit of virtual disaster recovery is that it minimizes the amount of configuration changes admins need to make in an actual DR scenario. In traditional or physical DR setups, IT teams often have to consult a playbook to understand what configuration changes to implement. This process is time-consuming and stressful in real DR scenarios, which means errors easily occur. However, with virtual DR systems, IT admins can configure failover IP address data, domain name systems and other items upfront, when they set up DR protection groups. Protection groups are collections of VMs -- such as all the VMs in a specific application stack -- that fail over concurrently and to the same point in time.

Virtual DR also enables rollbacks in small-time increments -- as small as a few minutes -- to find the most appropriate point-in-time copy of the VM. This capability would be useful, for example, if a group of live virtual machines was infected with ransomware and admins were unsure when, exactly, the attack occurred. When replication is based on logical unit numbers (LUN), such tools don't exist or are cumbersome to use.

Compared to physical setups, virtualized environments provide additional DR features. To take advantage of them, however, IT teams must lean on automation.

Invest in automation

Compared to physical setups, virtualized environments provide additional DR features. To take advantage of them, however, IT teams must lean on automation.

For example, virtualized environments make it easier for admins to perform rolling tests to ensure virtual machines spin up as expected in a DR scenario. These tests increase confidence in an organization's DR capabilities. Such functionality would be impractical in a physical DR environment, due to the overhead involved in continuous testing.

These testing capabilities, however, aren't native to virtual DR software. To perform rolling tests, admins must invest in automation tool sets, build automated processes and playbooks, and have a solid understanding of APIs.

These automated tests can, however, introduce some risk. For example, admins must be careful not to misconfigure test machines to come up in a live environment.

Don't mix the old with the new

For reasons related to cost or convenience, administrators sometimes leave physical components in an otherwise all-virtual DR environment. This is not a good practice, as it negates the benefits -- such as speed and efficiency -- that virtual DR delivers.

For example, the use of SQL Server log shipping -- or sending log backups from a primary database on a primary server to secondary database on a secondary server -- requires more hands-on DR management. SQL admins get involved and must spend time on proper configurations. Comparatively, failovers in all-virtual environments occur more quickly. And while virtualization does introduce licensing and resource costs, the convenience and speed of virtual DR setups outweigh them.

Dig Deeper on Disaster recovery facilities and operations

Data Backup