We are happy to announce that the paper “MulticoreBSP for C: A High-Performance Library for Shared-Memory Parallel Programming” by Albert-Jan N. Yzelman, Rob H. Bisseling, Dirk Roose and K. Meerbergen is published online in the International Journal of Parallel Programming, August 2013 (DOI 10.1007/s10766-013-0262-9). Enjoy !
Arrow Street is a C++11 library to create containers of objects with a “structures of array” and “arrays of structures of arrays” memory layout. Such layouts are known to be beneficial for better exploitation of SIMD instructions (SSE2, AVX, etc.). It is now available as an open source project on GitHub with a Google Group [...]
On September 22, the Sniper team will hold a tutorial at the IISWC conference, in Portland, Oregon. We will talk about how to use and extend the simulator, and give an overview of the new features in our upcoming 5.2 release. See here for more information.
We are pleased to announce that we have been granted a PRACE Preparatory Access Grant for doing scaling tests of Helsim, a 3D explicit electromagnetic particle-in-cell simulation with in-situ visualization developed in the ExaScience Lab. The grant allows for 200.000 core-hours on the Tier-0 Marie-Curie system in Paris. Happy running, Roel, Giovanni, Tom and Petar [...]
We are pleased to announce that the Intel European Exascale Labs Annual Report 2012 is now available, reporting on the activities of the Intel-BSC Exascale Lab in Barcelona, the Intel ExaCluster Lab in Jülich, the Intel ExaScale Computing Research Lab in Paris and of course our own Intel ExaScience Lab in Leuven. Happy reading !
We are happy to announce an update to the MulticoreBSP for C software. Highlights are: A new BSP primitive: the bsp_hpsend, Improved communication speed, Improved synchronisation speed (on machines with a large number of cores), New possibilities for advanced control of thread affinities as required on NUMA architectures, Compilation support for Windows; … and see the changelog [...]
We are happy to announce a talk that will be given by Pascal Costanza (Intel) on “Semi-automatic SIMD-efficient data layouts for object-oriented programs” at the 20th International Conference on Computing in High Energy and Nuclear Physics (CHEP2013). The talk will present a novel technique, made possible by new C++11 capabilities, which bridges the gap and [...]
On Wednesday 26 June 2013, from 10:30 to 12:00, Prof. Joost Schymkowitz (VIB SWITCH Laboratory) will give a seminar titled “Computational methods for modelling protein folding and aggregation”. The talk will take place at imec, in the Imec Auditorium. Directions to imec can be found here. External people will have to sign-in at the Imec Reception. Attendance is [...]
Our paper “Sampled Simulation of Multi-Threaded Applications” was presented at the 2013 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS). Based on the importance of the problem of multi-threaded sampling, and the quality of our solution, the program committee has honored this paper with the ISPASS Best Paper Award! Read the full [...]
We are pleased to announce that we have been granted a PRACE Preparatory Access Grant for the project “Development of load-balancing schemes for a particle in cell solver, and an adaptive mesh-refined Maxwell solver“. The grant allows for 250.000 core-hours on JUQUEEN, Europe’s fastest Supercomputer. Happy running, Dirk, Bram, Tobias and Bart !
- Invited talk on elPrep at the (IB)² Seminar on March 20, 2015
- New results and software on HPC sparse computation published
- Videos from the “High Performance Computing (HPC) for Life Sciences” event now available
- Jörg Kurt Wegner (Johnson & Johnson) giving closing keynote at EuroQSAR 2014
- Tutorial on GASPI, the PGAS API