Parallel computation of Feynman loop integrals

The need for large numbers of compute-intensive integrals, arising in quantum field theory perturbation calculations, justifies the parallelization of loop integrals. In earlier work, we devised effective multivariate methods by iterated (repeated) adaptive numerical integration and extrapolation, a...

Full description

Saved in:
Bibliographic Details
Published in:Journal of physics. Conference series 2012-01, Vol.402 (1), p.12029-8
Main Authors: de Doncker, E, Yuasa, F
Format: Article
Language:English
Subjects:
Applications programs
Computation
Extrapolation
Field theory
Integrals
Mathematical analysis
Multivariate analysis
Numerical integration
Parallel processing
Perturbation
Perturbation methods
Physics
Quantum field theory
Quantum theory
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:The need for large numbers of compute-intensive integrals, arising in quantum field theory perturbation calculations, justifies the parallelization of loop integrals. In earlier work, we devised effective multivariate methods by iterated (repeated) adaptive numerical integration and extrapolation, applicable for some problem classes where standard multivariate integration techniques fail through integrand singularities. In repeated integration, the function evaluations for the outer integral are independent integral computations for the next lower level and can be distributed to threads in a multi-core parallelization. The distribution level determines the number of dimensions in the inner integral below it and thus the granularity of the computation. We discuss a multi-threaded implementation over the OpenMP Application Program Interface.
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/402/1/012029