Showing posts with label supercomputing. Show all posts
Showing posts with label supercomputing. Show all posts


Amazon cloud computing for HPC

ArsTechnica has a nice article about the suitability of the Amazon cloud for High Performance Computing. I had linked to two previous stories about HPC clusters built on Amazon EC2. In short, Amazon EC2 tends to be better suited to embarrassingly parallel code where there is not much inter-node communication. This leaves out applications like finite-element simulations and molecular dynamics simulations. They mention R Systems which specialize in renting out Infiniband-connected clusters: Infiniband offers low latency, high speed connections.


50k-core supercomputer on Amazon EC2

Remember this previous story about a 30k-core Amazon EC2 cluster? Well, that's been topped. ArsTechnica reports that a company called Schrödinger ran a 50k-core Amazon EC2 instance for 3 hours to work on a cancer drug research problem. The cost was about $4800/hour. Actual deployment was handled by Cycle Computing. Cycle Computing also deployed the 30k-core cluster in the previous story.


30k-core cluster built on Amazon EC2 cloud

Ars Technica reports on a company called Cycle Computing that has built a 30k-core cluster on Amazon EC2 cloud for a client. It costs $1279 per hour to run, including fees to Amazon and Cycle Computing.

The cluster ran for about 7 hours, doing a computation for a pharmaceutical company. It had 3809 compute instances, each with 7GB RAM making 30472 cores, 26.7 TB RAM, and 2PB of disk space.

In the past, Amazon has built a virtual cluster that got onto the TOP500 list.

Pretty amazing.