Cloud Computing
Amazon Web Services applies supercomputing service to its cloud
The new instance has 16 cores and 60.5GB of RAM, and is priced at £1.50 per hour
By Mikael Ricknäs | IDG News Service | Published 10:10, 16 November 11
Amazon Web Services (AWS) wants to attract more high-performance computing users to its cloud and has launched a public beta of Cluster Compute Eight Extra Large, its most powerful cloud service yet.
AWS' cloud for high-performance computing applications offers the same benefits as it does for other other applications: it eliminates the cost and complexity of buying, configuring and operating in-house compute clusters, according to Amazon.
Also in this channel
Virtualisation, Big Data and BYOD
Check out our Business IT Hub for opinions and briefings. Read more
Applications that are a good fit for Amazon's cloud-based clusters include physics simulations, seismic analysis, drug design, genome analysis, aircraft design, and a variety of business computing and analytics applications, it said.
Every Cluster Compute Eight Extra Large (CC2) instance has two Intel Xeon processors, each with eight hardware cores.
Each CC2 instance also comes with 60.5GB of RAM and 3.37TB of storage, and it communicates with other instances - or virtual servers -- in a cluster using 10 gigabit ethernet. The service is priced at £1.50 (US$2.40) per hour for an on-demand instance and users can choose between Linux or Windows Server 2008 R2 as their operating system.
Users can also choose to buy a reserved instance, priced at £2,621 ($4,146) plus £0.34 ($0.54) per hour when signing a one-year contract, or bid for time on the EC2 Spot Market, where costs can be a lot lower, according to Amazon.
Users who want a system that would qualify among the Top500 most powerful supercomputers in the world could can launch their own array of 290 CC2 instances and pay less than £632 ($1,000) per hour.
To show what its cloud can do, Amazon also put together a cluster that used 17,024 cores and came in 42nd place on the latest version of the Top500 list of the world's fastest supercomputers, which was released this week.
For now, the cluster instances are only available in Amazon's US East (Northern Virginia) region, but the company plans to add that option to other regions next year.
