As I look back at the last year, I noticed that interest among our customers in digital resiliency has really been growing at a surprising (or maybe it shouldn’t surprise us at all) rate. There are an increasing number of requests for help in creating a robust failover architecture that will help them both enhance performance and increase reliability and availability.
I’ll take this opportunity to share with you the 3 primary use cases we see among both new and long-time customers.
Use Case # 1. Active:Active Failover
A computational software company serving financial and credit insurance organizations turned to Oracle Dyn as their business expanded. When their services became more critical to their end users’ success, they established service level agreements (SLAs) that specified extremely limited down time.
The organization implemented a second data center that would serve as their backup service location. The new data center was designed to be online and live—and was configured to serve requests in the event of a primary site failure.
We helped them set up active failover to monitor the server at the primary site at one-minute intervals with a TTL value of thirty seconds on the IP address of the primary server. Leveraging this configuration ensured that the longest outage during a failover would not be greater than 90 seconds in total. This enabled the organization to continue to meet their uptime SLAs and continue to maintain a competitive advantage over their competitors.
Use Case #2. Active:Passive Failover
Having multiple data centers for redundancy is quite common and provides for replication of data at multiple locations. In the event that one data center is not available, you can fail over a second “cold” site. We recently worked with a company in the fitness industry that sells fitness equipment and gear on their website. If their data center is not accessible, their customers can’t place orders and will likely buy from a competitor.
When they turned to Oracle Dyn, another vendor provided this fitness company’s DNS. But that vendor was not able to deliver a seamless failover solution, and the customer wanted their failover to be automatic, with automatic notification whenever a failover occurred.
Setting the lowest possible TTL ensures a fast failover to their secondary data center. This enabled them to achieve their uptime objectives and configure a solution their clients won’t notice any difference as their secondary site comes online to serve the request.
Use Case #3. VPN Failover
Working with multiple ISPs adds additional redundancy and resiliency. When it’s critical that employees are able to log in via VPN, no company can afford to lose access because a provider isn’t reachable. Having multiple providers configured is the first step in addressing this problem, but then you’ll need to fail over from one provider to the other seamlessly.
Many of the customers with whom we have worked have hosted their DNS in-house before coming to Oracle Dyn. Their in-house solution could not provide the failover solution that they needed even with two ISPs in place, because it was a manual process to switch between the ISPs.
With Oracle Dyn’s active failover and setting a low TTL of 30 seconds, our customers can automatically serve the failover IP address within 90 seconds—enabling them to maintain uptime and VPN access, no matter which ISP is in use.
So, if this has started you thinking about how to fix potential resiliency gaps in your own infrastructure, find out more here.