Performance Optimization of Baryon-block Construction in the Stochastic LapH Method
Preprint
Abstract
Implementations of measurement kernels in high-level Lattice QCD frameworks enable rapid prototyping, but can leave hardware capabilities significantly underutilized. This is an acceptable tradeoff if the time spent in unoptimized routines is generally small. The computational cost of modern spectroscopy projects however can be comparable to or even exceed the cost of generating gauge configurations and computing solutions of the Dirac equation. One such key kernel in the stochastic LapH method is the computation of baryon blocks; we discuss several implementation strategies and achieve a 7.2x speedup over the current implementation on a system with Intel(R) Xeon(R) Platinum 8358 processors, formerly Ice Lake.Keywords
All Related Versions
This publication has 28 references indexed in Scilit:
- Elastic nucleon-pion scattering at mπ = 200 MeV from lattice QCDNuclear Physics B, 2023
- Towards grounding nuclear physics in QCDProgress in Particle and Nuclear Physics, 2021
- Interactions of two and three mesons including higher partial waves from lattice QCDJournal of High Energy Physics, 2021
- Two-nucleon -wave interactions at the SU(3) flavor-symmetric point with : A first lattice QCD calculation with the stochastic Laplacian Heaviside methodPhysical Review C, 2021
- Finite-volume energy spectrum of the systemPhysical Review D, 2020
- Three pion spectrum in the channel from lattice QCDPhysical Review D, 2020
- Three-body unitarity versus finite-volume spectrum from lattice QCDPhysical Review D, 2020
- CLS 2+1 flavor simulations at physical light-and strange-quark massesEPJ Web of Conferences, 2018
- Simulation of QCD with N f = 2 + 1 flavors of non-perturbatively improved Wilson fermionsJournal of High Energy Physics, 2015
- Unified contraction algorithm for multi-baryon correlators on the latticeComputer Physics Communications, 2013