FAQ Friday: “Active Active” Data Centers Are a Hot Hot Topic Within Disaster Recovery

By Brian Schwartz, TierPoint Content & Social Media Marketing Manager

This week I spoke with Christian Lappin, a TierPoint senior sales engineer, who is a business continuity/disaster recovery expert and an active member of the FBI Infragard Task Force. He speaks with clients each week about high availability, and says that in recent weeks he has been receiving lots of questions about “active active” data centers and how clients can transition into one from their current environment.

He says that the term “active active” refers to both data centers and applications. The idea is to have data flowing into two locations for better load balancing and automatic failover should one of the locations go down. When you have active active data centers working in concert with “active active” applications you get what’s called continuous availability.

Continuous availability is particularly important in scenarios where there are lots of real-time transactions and when unplanned downtime causes chain reactions that negatively impact downstream services such as disrupting customers’ own customer’s applications.

To create an “active active” data center pair, “Sites must be connected by links that can support latency less than about 5 milliseconds,” Lappin said. “Depending on the actual data transport method, this could translate to about 60 miles apart.”

It is possible to set up an “active active” data center within a single location for high levels of availability, but you run the risk of a natural disaster or other threat taking down everything at once.

Lappin says that before customers invest in an “active active” data center, they have to consider a number of factors including their current environment and how suitable it would be to convert into an “active active” data center. Also, important, they have to consider “their tolerance for downtime,” Lappin added. Part of the process that Lappin goes through to help clients make that determination is to help them evaluate their recovery time objective (RTO) and recovery point objective (RPO). Essentially that means that there are services that can recover data at different speeds according to need and budget.

Click on the audio clip to hear more insights from Christian Lappin about “active active” data centers. Have a question, leave it in the comments, and he will respond.

Connect with Brian.