Hybrid systems and services that fully leverage the performance of accelerators
Although the emergence of multi-core CPUs has allowed manufacturers to keep pace with the steadily growing demand for more computing power, energy consumption, space, and cooling have become major inhibitors to computing systems expansion. Hence the success of accelerators such as GPUs (Graphics Processing Units) offering breakthrough performance plus outstanding space and energy efficiency. Derived from technologies initially developed for game machines, GPUs are highly parallel processors designed to seamlessly boost High Performance Computing applications without impacting power or cooling infrastructures.
Implementing GPU computing is getting easier every month as new tools become available, but it is still not trivial. To help you innovate with GPUs with maximum payoff, Bull offers complete GPU-based solutions:
![]() |
Developed by the Bull R&D teams, the bullx B505 ultra-dense, double-width blades combine two Intel® Xeon® processors and two new-generation NVIDIA® Tesla™ GPUs. They can be associated with B500 compute blades within the same bullx chassis. The integration of the GPUs inside the blades maximizes performance. The bullx B505 blades are the only blades on the market designed to provide maximum bandwidth between each GPU and its host CPU, thanks to two chipsets and a double interconnect on each blade:
|
Depending on the type of application and its degree of optimization, GPUs can accelerate processing by a factor of 1 to 100! Expertise is therefore essential to get the most out of GPUs. Bull has a long experience of systems associating GPUs and CPUs, and places this distinctive expertise at the service of Extreme Computing customers. Bull engineers have had hands-on experience of porting and optimizing applications for GPU computing through many cooperative projects with customers, such as a cluster with 46,080 GPU cores, designed by Bull for GENCI. Bull experts deliver end-to-end services, from system design to advanced user training, through application optimization for GPUs.
In the last few years, Bull has designed for its customers accelerator-based systems of all sizes, from test systems with a few GPUs – purchased by several universities for example – to very large hybrid systems such as those developed for GENCI and Petrobras. Bull helps its customers optimize their GPU utilization, with very varied codes such as BIGDFT (nano-sciences), SPECFEM3D (seismic), NAMD (biochemistry), but also for oil and gas research with the porting of RTM (Reverse Time Migration) codes to GPU.
The bullx supercomputer suite integrates all the tools needed to operate a hybrid system comprising Intel® Xeon®-based servers and NVIDIA® Tesla™ GPUs. The task management environment supports the allocation of applications to the relevant computing resource. An application can naturally use both computing resources so as to fully capitalize on the rich potential offered by this type of system.
19/03/2012
Bull, CEA, F4E and JAEA inaugurate the Helios supercomputer in Rokkasho (Japan), dedicated to the Nuclear Fusion program
The supercomputer, delivering over 1.5 Petaflops, will provide the computer modeling and simulation capabilities needed for the 'Broader Approach' program, linked to the ITER initiative.
Read more >>
15/02/2012
CURIE speeds up European research
Capable of up to two million billion operations a second (or 2 Petaflop/s) GENCI's CURIE supercomputer - designed by Bull - opens up unprecedented new possibilities for academic and industrial research in Europe
Read more >>
November 2011
Special Edition of La Recherche - Supercomputers: at the frontiers of Extreme Computing
New horizons / Major Challenges / The future: exascale computing
Read more >>