Name: Future of Many-Core Computing in Science
Start: 2014-05-29T08:00:00+02:00
End: 2014-05-29T18:00:00+02:00
Location: Wigner Datacenter

Future of Many-Core Computing in Science

Thursday, 29 May 2014 - 08:00

Monday, 26 May 2014
Tuesday, 27 May 2014
Wednesday, 28 May 2014
Thursday, 29 May 2014

08:30 Registration - enter to the Campus
Registration - enter to the Campus
08:30 - 09:00
Room: 2nd Floor, Meeting Room
09:00 Opening -- Greetings from the General Director - Peter Levai (WIGNER RCP)
Opening -- Greetings from the General Director
- Peter Levai (WIGNER RCP)
09:00 - 09:15
Room: 2nd Floor, Meeting Room
09:15 Portable HPC for High-Performance Simulation - Federico Carminati (CERN)
Portable HPC for High-Performance Simulation
- Federico Carminati (CERN)
09:15 - 10:00
Room: 2nd Floor, Meeting Room High Energy Physics code has been known for making limited use of high performance computing architectures. Efforts in optimising HEP code on vector and RISC architectures have yield limited results and recent studies have shown that, on modern architectures, it achieves a performance between 10% and 50% of the theoretical one. Although several successful attempts have been made to port selected codes on GPUs, no major HEP code suite has a "High Performance” implementation for accelerators. With LHC undergoing a major upgrade and a number of challenging experiments on the drawing board, HEP has to try making the best usage of the new hardware. This activity is one of the foci of the SFT group at CERN, which hosts, among others, the Root and Geant4 projects. One important initiative is centred on the development of a high-performance prototype for particle transport, the GeantV project. A good concurrency level has been achieved by Geant4 with the parallelisation at event level. However, apart the sharing of data structures, this does not increase the number of events per second produced on a loaded system, which requires an efficient use of the low level instruction parallelism and pipelining of modern processors. In GeantV project we have implemented a framework that allows scheduling vectors of particles to an arbitrary number of computing resources in a fine grain parallel approach. Via template code specialisation we are able to recast simple computational kernels do different architectures, minimising code duplication. This approach is already providing very promising results in the treatment of the detector geometry and we plan to extend it soon to the optimisation of the physics code. The talk will review the current satis and the perspectives of the development of a simulation framework able to profit best from the recent technology evolution in computing.
10:00 The Future Computing with Silicon Computer Systems - Gábor Lehoczki (Silicon Computers LTD)
The Future Computing with Silicon Computer Systems
- Gábor Lehoczki (Silicon Computers LTD)
10:00 - 10:20
Room: 2nd Floor, Meeting Room
10:20 Coffee Break
Coffee Break
10:20 - 10:40
Room: 2nd Floor, Meeting Room
10:40 AMD Heterogeneous System Architectures : Software and Hardware Ecosystem - Dmitry Kozlov (Advanced Micro Devices INC.) Bruno Stefanizzi (Advanced Micro Devices INC.)
AMD Heterogeneous System Architectures : Software and Hardware Ecosystem
- Dmitry Kozlov (Advanced Micro Devices INC.)
- Bruno Stefanizzi (Advanced Micro Devices INC.)
10:40 - 11:50
Room: 2nd Floor, Meeting Room
11:50 Bi-directional particle tracing on the GPU with applications in PET - László Szirmay-Kalos (Technical University of Budapest)
Bi-directional particle tracing on the GPU with applications in PET
- László Szirmay-Kalos (Technical University of Budapest)
11:50 - 12:10
Room: 2nd Floor, Meeting Room
12:10 Trends in Visualization and Physics in Real-Time Graphics - Pál Mezei (Solid Angle LTD)
Trends in Visualization and Physics in Real-Time Graphics
- Pál Mezei (Solid Angle LTD)
12:10 - 12:30
Room: 2nd Floor, Meeting Room
12:30 Lunch Break
Lunch Break
12:30 - 13:30
Room: 2nd Floor, Meeting Room
13:30 QCD on Lattice - Sándor Katz Ferenc Pittler (Eötvös Univeristy, Budapest)
QCD on Lattice
- Sándor Katz
- Ferenc Pittler (Eötvös Univeristy, Budapest)
13:30 - 13:50
Room: 2nd Floor, Meeting Room In numerical simulation of quantum chromodynamics the presence of large number of integration variables requires up-to-date hardware technologies, and software developed for the particular hardware. Due to the locality in field theory the main source of performance improvement nowadays is parallelization and simulation on GPU clusters. We present the GPU cluster at the Eotvos Lorand University. We briefly review the Krylov-Schur algorithm (which we use for matrix diagonalization and matrix function computation) and compare the performance of the GPU and parallel CPU implementation.
13:50 GPU Code-generation for Differential Equation Solvers - Dániel Berényi (Wigner RCP of the HAS, Eötvös University)
GPU Code-generation for Differential Equation Solvers
- Dániel Berényi (Wigner RCP of the HAS, Eötvös University)
13:50 - 14:10
Room: 2nd Floor, Meeting Room
14:10 GridRipper >> Lattice Template Library - Máté Ferenc Nagy-Egri (WIgner RCP of the HAS)
GridRipper >> Lattice Template Library
- Máté Ferenc Nagy-Egri (WIgner RCP of the HAS)
14:10 - 14:30
Room: 2nd Floor, Meeting Room
14:30 Accelerated Monte Carlo Particle Generators for the LHC - Gergely Gábor Barnaföldi (Wigner RCP RMI of the Hungarian Academy of Sciences)
Accelerated Monte Carlo Particle Generators for the LHC
- Gergely Gábor Barnaföldi (Wigner RCP RMI of the Hungarian Academy of Sciences)
14:30 - 14:50
Room: 2nd Floor, Meeting Room
14:50 Coffee Break
Coffee Break
14:50 - 15:10
Room: 2nd Floor, Meeting Room

15:10 Closed Session
Closed Session
15:10 - 17:00
Room: 2nd Floor, Meeting Room