With the rise of both man-made and natural disasters (including fires and earthquakes), the disaster recovery (DR) market has growing importance in protecting an enterprise and its user community, according to RackWare co-founder and CEO Sash Sunkara.
“It is a really critical requirement [for enterprises] and when people think about DR there are certain things that they should really think about when putting a plan together,” Sunkara said. This includes implementing the correct infrastructure monitoring and disaster recovery toolset so that an enterprise’s network, applications, and end-users don’t experience any downtime.
Sunkara recommends having a plan and a tool implemented that includes prioritization and automation and being able to easily test your tools and network. This tool should include monitoring of some kind, she says. At the very least a company’s monitoring tool should enable seamless integration between DR and the monitoring servers.
“Monitoring servers have numerous benefits for constructing a DR plan,” Sukara explained. “One is the ability to right-size servers in the DR site. For any servers overprovisioned on the origin, right sizing on the target can [lead to] substantial cost savings. Understanding application affinities helps to group servers into consistent policies for syncing and failover. Monitoring also helps identify high transaction periods of time where it may be beneficial to pause syncs until more resources (both compute and network) are available.”
Making a Plan
In identifying priorities it’s important first to understand the geographic risks associated with your data center and branch locations, says Todd Matters, co-founder and chief architect at RackWare.
“DR is a lot more than just backup,” he said. “Backup is making sure that you have a copy of something, but DR means that you’re protecting from certain specific events and whatever those event risks are … there are different threats [based on] geography and how far away your DR site needs to be is a very important consideration.”
Then, it is also important to understand the needs of your company — which includes the different application and server priorities, Matters added. Your DR plan should also account for your necessary recovery time and how much money you’re willing to spend.
One of the things that companies don’t often think about is that in the case of a real event, your employees will be busy, which is where automation comes in.
“With things like the lives of their families, they’re not going to be thinking about their data center or their applications. They need to be assured that all of that is going to be up and their customers are going to be okay for them having to have a hundred people from their organization dealing with this,” said Sunkara.
Whatever tool you choose should have automation built in so that your DR plan can be executed with little human intervention.
Regardless of the DR software or provider that an enterprise selects, it’s vital that it enables you to easily test your plan to make sure that it’s a good one.
“If you can do a frequent test, easily, cost-effectively taking a short amount of time, then you have more confidence and you reduce the risk of your organization,” Sunkara said.
One of the biggest mistakes that RackWare sees is that companies don’t test for fallback. Sunkara explained that fallback refers to the process after the DR plan has been executed and applications have been moved to a different data center (or DR site) when an enterprise must return its applications to the primary data center.
“No one ever tests that and if you don’t get that to work it’s gonna cost you an unbelievable amount of money and personnel and time to deal with that,” Sunkara said. “So when you do a DR test you should not only think about failing over to one environment but also falling back to your primary environment.”
RackWare is a Fremont, California-based software company that focuses on cloud management, including hybrid-cloud management.
According to Sunkara, its main product lines encompass three areas: migration, which includes moving applications to the cloud; disaster recovery and backup resiliency; and managing your cloud resources, which includes moving between different cloud providers.
Sunkara noted that RackWare has a number of competitors that focus on its different areas (DR, resource management, and migration), the company differentiates itself by having all three.
“You don’t have to go to one player to move your applications to the cloud, you don’t have to go to a different player to protect those workloads, and then a third player to be able to manage those workloads in the cloud,” she said, adding that RackWare is also one of the few players that can give broad cloud coverage to the enterprise rather than supporting a single cloud.