Lattice2010

Session

Parallel 54: Algorithms and machines

18 Jun 2010, 16:40

Villasimius, Sardinia

30. GPU Performnace of Conjugate Gradient Solver with Staggered Fermions

Hyung-Jin Kim (Seoul National University)

18/06/2010, 16:40

Algorithms and machines

The progress in GPUs (graphic processing units) is much faster than that of CPUs. We use GPUs of nVIDIA GTX 295, GTX 285, and GTX 480 models to probe the computing performance of the conjugate gradient solver for the Dirac equations made of staggered fermions. We use MILC fine lattices (283 × 96). The performance test is done using 4 GPUs. We use CUDA v2.3 and v3.0 libraries to compile the...

37. GPU-Based Conjugate Gradient Solver for Lattice QCD with Domain-Wall Fermions

Yao-Yuan Mao (National Taiwan University)

18/06/2010, 17:00

Algorithms and machines

We present a GPU-based conjugate gradient (CG) solver for lattice QCD with domain-wall fermion (DWF). It is well known that CG is the most time-consuming part in the hybrid Monte Carlo simulation of unquenched lattice QCD, which becomes even more computational demanding for lattice QCD with exact chiral symmetry. We have implemented the CG for a general 5-dimensional DWF operator on NVIDIA's...

38. Implementation of the Neuberger overlap operator in GPUs

Bjoern Walk (Institut für Kernphysik, Universität Mainz)

18/06/2010, 17:20

Algorithms and machines

To describe a decay like $K \to \pi\pi$ a formulation of lattice QCD has to be found which respects chiral symmetry. For this, the Dirac operator has to fulfill the Ginsparg-Wilson relation and one particular solution to this is the Neuberger overlap operator. The downside of having a chiral formulation of QCD on the lattice is that its computational costs are orders of magnitudes higher. On...

72. Domain decomposition method on GPU cluster for Lattice

Yusuke Osaki (Hiroshima University, depertment of physical science)

18/06/2010, 17:40

Algorithms and machines

The modern graphic processing unit (GPU) has become one of the powerful computational device. Applying GPU to the lattice QCD computation has been investigated especially for the inversion solver of the discretized Dirac operators. Most of the previous works on lattice QCD with GPU, however, have been done with single GPU card. Since the single GPU performance and the memory size are...