Amazon installs first European HPC cluster in Ireland
Utility supercomputing in the cloud comes closer for AWS European customers
By Sophie Curtis | Techworld | Published: 12:55, 26 June 2012
Amazon Web Services (AWS) has installed a cluster system at its EU West facility in Ireland, allowing customers in the region to run high performance computing (HPC) instances in the cloud.
The Cluster Compute Eight Extra Large (cc2.8xl) instances went live on Friday 22 June, the company said in a blog post. This is the first time that AWS’s HPC cloud technology has been available outside of the US.
Each instance includes a pair of Intel Xeon processors, 60.5GB of RAM, and 3.37TB of instance storage, according to AWS. Each processor has 8 cores and hyperthreading is enabled, so users can execute up to 32 threads in parallel.
Related Articles on Techworld
The instances are connected to a 10GbE network, and can be grouped together for low-latency connectivity to other CC2 instances.
The new HPC cluster will appeal to organisations that are working with large, complex datasets, including financial services, oil and gas, life sciences, social gaming, advertising, e-commerce and media, according to Matt Wood, product manager for high performance and data intensive computing at Amazon Web Services.
“The arrival of the cc2.8xlarge instance size in the EU West region allows customers who store data in that geography to compute, analyse, ask questions and find insight from that data using high performance Intel Xeon E5 processors,” said Wood.
“This brings the power of a purpose designed HPC environment with 10GbE fully bisectional networking to customers with the same on-demand, utility metering offered by other AWS services, providing faster turn arounds, larger computational runs and a shorter time to market.”
The CC2 instances in Ireland are priced at $2.70 (£1.73) per hour for Linux and $2.97 (£1.90) per hour for Windows, said Amazon.
Speaking at the launch of Intel's Xeon E5 processor family earlier this year, AWS technology evangelist Dr. Matt Wood said that we are now entering the era of utility supercomputing, whereby anyone can dial up computational resources and massive storage requirements on the fly.
He said that scientific and financial organisations with massive computational demands will increasingly be able to rent resources from the cloud to be able to do their work – whether it happens to be product modelling, simulation or informatics – without having to invest in huge infrastructure.
“Traditionally these organisations would have to provision for 10-15 percent over the peaks in demand, but the cloud allows for bursty scalability, lowering the barriers to entry and allowing them to spend at least 70 percent of their time on differentiated work, rather than keeping the lights on,” he said at the time.
Amazon said that cc2.8xl instances had formed the basis of a computing cluster which clocked a maximum speed of 240 teraflops. The cluster, which contains 17,024 cores and 65.968 TB of RAM currently ranks number 72 on the Top500 list.