The Hybrid Processing Units for Science (HPU4Science) team aims at tackling difficult scientific problems by developing novel computational methods that make good use of the HPU, Hybrid Processing Units, i.e. CPUs and GPUs.
Our current cluster is a group of six desktop-type computers powered by Intel i7 or Core 2 Quad processors together with GPUs that range from the GTX 280 to GTX 590s, for a theoretical peak performance of 34.5 TFLOPS. The cluster is built by and located in the Laboratoire de Chimie de la Matière Condensée de Paris (LCMCP), a mixed CNRS/Paris6 chemistry lab located at Chimie ParisTech that specializes both in the development of new materials and in the understanding and characterization of existing ones.
With this web site, you can learn more about us and the work we are doing. If you like it then you'll probably want to work on our projects and have a look on the job opportunities.
Recent News
-
Upgrades on all the machines of the cluster
We just finished updating the NVIDIA drivers, CUDA and PyCUDA on workers and master.For the NVIDIA, driver it's the version 290.10.For CUDA, we use the CUDA ...
Posted Feb 7, 2012 3:17 PM by Yann Le Du
-
The GTX 295 is dead
Few days ago, we moved the GTX 295 in another machine, and since it dead. It worked well since the 25th of June 2009 until it was movec to another ...
Posted Feb 8, 2012 5:46 AM by El Afrit Mariem
-
The HPU4Science Core in Tunisia for five days
The Core of Hpu4Science will be in Tunisia from 11th to 15th of January to start a new collaboration between the LCMCP laboratory under the direction of Pr. Didier Gourier ...
Posted Dec 26, 2011 2:52 AM by El Afrit Mariem
-
Stage 2012 au sein de l'équipe HPU4Science
We are again looking for one or two enterns in the HPU4Science team.We are looking for three types of profiles: mathematically oriented, application programming oriented or system administration oriented
Posted Dec 21, 2011 2:58 AM by El Afrit Mariem
-
Parallel Sparse Linear Algebra, INRIA, Bordeaux
We are following a special course on parallel sparse linear algebra this whole week. Our goal is twofold:meet specialists in the field and see if we can discover some ...
Posted Nov 29, 2011 12:26 AM by El Afrit Mariem
|
|
