With Hurricane Sandy expected to hit the US East Coast as one of the worst storms in the region’s history, residents and businesses along with Presidential candidates are scrambling to prepare for the worst. Even non-local businesses and entities that are reliant on Amazon Web Services’ US East Region are concerned about being affected by outages resulting from the storm.
One AWS user asked on the AWS fourm, how to ensure 0% downtime and here is what AWS had to say:
There Is No Anti-Outage Magic Pill
Another participant (named Skaperen) on the AWS forum elaborated on the official AWS response and made some great points. Here is a brief summary:
- The cloud is not a magical place – failure can and does happen, just like anywhere else.
- All IaaS can do is provide you with affordable tools for creating redundancy – but it’s up to you how to do it.
- There is no single “right way” to do redundancy. It all depends on your unique needs – once again you are in the driver’s seat.
- “Build-it-yourself’ duplication is expensive, so using AWS’ cloud offerings in multiple regions (to the extent that you require) is definitely the way to go for most businesses
Going Redundant – Some Basics
In the most general terms, redundancy is something which really needs to be planned for. One of the first steps is to make sure your applications have been designed properly for redundancy. Once your architects have taken care of that, you can begin to consider how much redundancy your business requires and how much redundancy you can afford to pay for.
The ultimate would be working with multiple cloud vendors in different regions (perhaps in addition to your own on-site data center), but running your application/site etc. in multiple AWS regions should also offer excellent insurance against outages. And finally, working with multiple availability zones within a single AWS region, should provide a good level of redundancy in most normal circumstances.
Discover how Cloudyn can help you optimize your cloud – Sign-Up for FREE Now!