Multi-mass solvers for lattice QCD on GPUs

Graphical Processing Units (GPUs) are more and more frequently used for lattice QCD calculations. Lattice studies often require computing the quark propagators for several masses. These systems can be solved using multi-shift solvers but these algorithms are memory intensive which limits the size of...

Full description

Saved in:
Bibliographic Details
Published in:Journal of computational physics 2012-02, Vol.231 (4), p.1866-1878
Main Authors: Alexandru, A., Pelissier, C., Gamari, B., Lee, F.X.
Format: Article
Language:English
Subjects:
Algorithms
BiCGstab
BiCGstab-M
Computation
Computational techniques
Exact sciences and technology
Fermions
GPU
Lattices
Mathematical analysis
Mathematical methods in physics
Physics
Quarks
Solvers
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Graphical Processing Units (GPUs) are more and more frequently used for lattice QCD calculations. Lattice studies often require computing the quark propagators for several masses. These systems can be solved using multi-shift solvers but these algorithms are memory intensive which limits the size of the problem that can be solved using GPUs. In this paper, we show how to efficiently use a memory-lean single-mass solver to solve multi-mass problems. We focus on the BiCGstab algorithm for Wilson fermions and show that the single-mass solver not only requires less memory but also outperforms the multi-shift variant by a factor of two.
ISSN:0021-9991
1090-2716
DOI:10.1016/j.jcp.2011.11.003