Amazon outage hits the big names
A report from Spiceworks shows how the popularity of Amazon Web Services underpins some of the biggest internet names. The AWS storage service – S3 – experienced a series of failures on February 28th in their East Coast installations. The precise cause of the Amazon S3 outage is not clear but the impact was felt for almost 3 hours and affected some of the biggest online brands including:
Wix – provider of hosted WYSIWYG-designed websites for over 90 million customers
Twitch – Amazon’s video streaming service, a huge hit and revenue-earner for the gaming community
Docker – the open-source container management solution, adopted by millions of IT operations
Expedia – the premier travel booking site
IFTTT – free-to-use technical event flow-and-process solution for real-time processing
Murdoch’s News Corp
Mailchimp – leading email distribution list management solution
Medium– blogging and social networking site
Slack – business-orientated messaging and collaboration
Trello – community-based project management and visualisation
The US Securities and Exchange Commission (SEC)
Not an exhaustive list, more details have been provided by TechCrunch, VentureBeat and theVerge – serving to illustrate how cloud services underpin the modern internet. This is not a warning against the use of AWS or the wider cloud but the incident does demonstrate the need for effective contingency planning. Adopters might assume that the cloud guarantees availability and eliminates the need for resilience in service design. This is clearly not the case. AWS – and others – do offer geographical distribution and replication of data but, naturally, at additional cost.
With high-profile migrations to public cloud providers – such as Netflix moving their entire streaming platform to AWS – elastic platforms remain an excellent way to eliminate complexity and cost. Prototyping, proofs-of-concept and service expansion are also important use cases.