Amazon Web Services (AWS) Outage: 3 Key Lessons Learned
It's no secret that Amazon Web Services (AWS) is a leader in the cloud space - in fact, according to Gartner, it's THE leader in the public cloud storage market. But even giants falter sometimes. A massive outage on Feb. 28 at Amazon's S3 storage service provider caused technical issues that were felt across the globe and resulted in downtime for many companies and individuals. The outage began at approximately 12:30 p.m. EST and lasted for nearly four and a half hours, affecting both front-end and back-end operations for countless websites. For the first two hours, Amazon couldn't update its own service health dashboard, since it was hosted on AWS.
Amazon's Simple Storage Service (S3), one of AWS' data storage options, is used by approximately 148,213 websites and 121,761 unique domains. Internet of Things (IoT) devices were also affected by the issue, leaving many individuals with no remote access to smart devices like thermostats, lighting, and even security cameras. It was reported that the outage was simply the result of human error. An Amazon team member attempting to debug a billing system incorrectly typed a command that impacted a large number of servers.
Platforms like AWS give businesses a competitive edge by allowing them to deploy and host applications in the cloud. One of the most appealing benefits of cloud storage solutions is high availability, but last week's incident begs the question: what happens when one of the world's top cloud leaders does go down and how does that impact business continuity for its thousands of customers?
iCorps' VP of Technology, Jeffery Lauria, explains three lessons on business continuity in the cloud:
As Jeffery explains in the video above, expecting cloud providers to provide 100% uptime is unrealistic. Here are three lessons to keep in mind to avoid downtime in the cloud:
- Consider a hybrid cloud approach. Relying on the availability of one cloud provider will mean that, at some point, you'll be at the mercy of that provider's ability to get your applications, systems, or website back up and running. Strategically distributing your systems across cloud providers can help you avoid that situation.
- Consider a geographically dispersed cloud approach. The AWS incident effected customers on the East Coast. When you choose cloud services that are geographically dispersed, or use a service that replicates your data securely to multiple data centers in diverse locations, you can have confidence that a localized interruption won't impact your business. (Read: Why Geography Matters in Cloud Backup)
- Don't forget about data backup and disaster recovery in the cloud. Disasters happen - even in the cloud. When you move services, applications, data or systems to the cloud, you should not assume that they are automatically backed up, indestructible or eternal. Make sure you have a disaster recovery and business continuity plan in place that is tested frequently.
Hyperscale public cloud providers such as Amazon Web Services (AWS) and Microsoft are dominating the market for public cloud storage. Their innovation, scale, rate of growth, and security standards will make it difficult for other cloud vendors to keep up. Despite this hiccup for AWS, the provider will remain a top contender in the cloud space. As a Forbes technology writer points out, AWS up-time was still ~99.59% (only .36% beneath their target) the day that the media claimed "AWS broke the internet." We would be surprised, however, if this incident didn't draw attention to AWS' close competitor, Microsoft Azure - allowing Microsoft to slim the gap. Azure's flexible computing power also allows businesses to host websites on a secure cloud server that is managed by Microsoft and backed by their uptime guarantee, providing companies and their customers with reliable web hosting services.
No matter which cloud platform your business is considering, it's important to conduct as much due diligence before tying the knot with a provider. From questions on downtime to data security, you should feel confident that the solution you choose is the right fit for your business needs, expectations and requirements. Just because every workload or business application could go to the cloud, doesn't mean that it should. Can your business afford for its website to go down or have a glitch for even an hour? Maybe, maybe not. A Managed Services Provider with cloud expertise can help you determine those needs and expectations and help you select the cloud - or hybrid cloud - solution that works best for your business.
To stay informed on IT outages like the Amazon Web Services (AWS) Outage, subscribe to the iCorps blog! If you're in the process of researching cloud providers, you might want to check out: 7 Data Security Questions to Ask Your Cloud Provider.