Amazon Cloud Downtime Raises Questions
April 26th, 2011 By:THN News
(The Hosting News) – Last week’s server downtime of Amazon Web Services has raised questions into the overall reliability of the usually exceptional stable cloud computing. The incident has also lead to more calls for transparency from cloud-based providers.
Amazon Web Services, which provides hosting to popular internet sites such as Reddit, Quora, Hootsuite and services similar to Relational Database (RDS) and Elastic Compute (EC2) began suffering downtime on Thursday of last week. By Monday, Amazon said most of the services that rely on their cloud-based computing platform were back to normal.
Despite the somewhat return to normalcy, Amazon has stated through their Web Services Health Board that a small number of volumes were lost during the server failure. Amazon is currently communicating with these clients over the issue.
Meanwhile, in the aftermath of the crash, Amazon has also promised a full analysis of the incident and stated, “We are digging deeply into the root causes of this event and will post a detailed post mortem”.
While Cloud-Computing is known in large part for its reliability and its ability to cut costs, some analysts wonder what the incident means for the viability and future of cloud-computing in general.
Ken Brill of the Uptime Institute, which provides data research noted that despite the overall reliability of Cloud, the computing service forces Cloud users to focus processing on “fewer, bigger places”. Brill stated the end result of this was that when something does fail, it tends to happen on a much larger scale.
Other analysts also note that the incident shows that blind-faith should not be placed in Cloud just yet as the platform in still in its infancy. Cloud has recently seen an upsurge in popularity, especially within government agencies who are working to implement the platform into their daily activities.
|Corero Network Security|
* Real-time DDoS protection for you and your customers
* Purpose built DDoS protection appliance deployed at the Internet edge
* On-premises technology is designed to handle volumetric network-based DDoS attacks or floods, reflective and amplified spoof attacks, like DNS and NTP attacks, as well as application layer attacks - such as slow loris, slow read etc
* Unique, slim-line appliance family delivers 10 Gbps full-duplex performance in a 1/4 wide, 1 RU form factor, enabling Hosting providers to deploy a combination of SmartWall TDS appliances to deliver the performance, connectivity and security required
* Providers are now enabled to offer creative new offerings, such as DDoS and cyber threat protection, enhanced security SLAs as well as visibility and reporting through an analytics portal that can be leveraged as a premium service offering