J Supercomput
DOI 10.1007/s11227-014-1207-9
FuPerMod: a software tool for the optimization
of data-parallel applications on heterogeneous platforms
David Clarke · Ziming Zhong · Vladimir Rychkov ·
Alexey Lastovetsky
© Springer Science+Business Media New York 2014
Abstract Optimization of data-parallel applications for modern HPC platforms
requires partitioning the computations between the heterogeneous computing devices
in proportion to their speed. Heterogeneous data partitioning algorithms are based on
computation performance models of the executing platforms. Their implementation
is not trivial as it requires: accurate and efficient benchmarking of computing devices,
which may share resources and/or execute different codes; appropriate interpolation
methods to predict performance; and advanced mathematical methods to solve the
data partitioning problem. In this paper, we present FuPerMod, a software tool that
addresses these implementation issues and automates the development of data parti-
tioning code in data-parallel applications for heterogeneous HPC platforms.
Keywords Heterogeneous computing · Data partitioning ·
Computation performance models
1 Introduction
Many scientific applications implement data-parallel algorithms, originally designed
for homogeneous HPC platforms. The applications range from linear algebra routines
to computer simulations, such as computational fluid dynamics. Efficient execution of
This research is supported by Science Foundation Ireland (Grant 08/IN.1/I2054).
D. Clarke · Z. Zhong · V. Rychkov (B ) · A. Lastovetsky
School of Computer Science and Informatics, University College Dublin, Dublin, Ireland
e-mail: vladimir.rychkov@ucd.ie
A. Lastovetsky
e-mail: alexey.lastovetsky@ucd.ie
URL: http://hcl.ucd.ie/project/fupermod
123