Yesterday, Amazon reported high error rates with S3 (simple storage service) in its US-East-1 region, which impacted its other Amazon Web Services (AWS). Translation: Internet Outage. USA Today reported this outage affected millions of companies across the United States – of every size imaginable.
When something like this happens, everyone starts rethinking their decision to put mission-critical workloads in a public cloud. Here are our thoughts on what the AWS outage should leave you considering:
- You have to consider business continuity as it relates to your needs, as well as the associated economics. There are two sides to this question of downtime that may pull you in opposite directions: A) How well do my customers tolerate downtime? B) How much time and how many resources am I willing to invest to avoid downtime? The correct answer then becomes a new question altogether: What is business continuity worth to me, my business and my customers?
- Consider using multiple regions to improve your resiliency and redundancy. Beyond regions, think about overall portability with options like a hybrid cloud, or even redundant dedicated servers.
- The most important thing to consider is if you have a trusted advisor working with you on your cloud solution to map out possible outage scenarios and to help you arrive at the right infrastructure decisions for your customer expectations and budget.
Common sense? Sure. Many of you have likely arrived at these same conclusions over the years. But when your public cloud provider is down, and with it, your customers’ infrastructures and livelihoods, common sense can go right out the window.
Let’s make this a dialogue. What lessons did you take away?