Automatically Harnessing Sparse Acceleration
School of Informatics, The University of Edinburgh, United Kingdom
arXiv:2001.07938 [cs.PF], (22 Jan 2020)
@misc{ginsbach2020automatically,
title={Automatically Harnessing Sparse Acceleration},
author={Philip Ginsbach and Bruce Collie and Michael F. P. O’Boyle},
year={2020},
eprint={2001.07938},
archivePrefix={arXiv},
primaryClass={cs.PF}
}
Sparse linear algebra is central to many scientific programs, yet compilers fail to optimize it well. High-performance libraries are available, but adoption costs are significant. Moreover, libraries tie programs into vendor-specific software and hardware ecosystems, creating non-portable code. In this paper, we develop a new approach based on our specification Language for implementers of Linear Algebra Computations (LiLAC). Rather than requiring the application developer to (re)write every program for a given library, the burden is shifted to a one-off description by the library implementer. The LiLAC-enabled compiler uses this to insert appropriate library routines without source code changes. LiLAC provides automatic data marshaling, maintaining state between calls and minimizing data transfers. Appropriate places for library insertion are detected in compiler intermediate representation, independent of source languages. We evaluated on large-scale scientific applications written in FORTRAN; standard C/C++ and FORTRAN benchmarks; and C++ graph analytics kernels. Across heterogeneous platforms, applications and data sets we show speedups of 1.1x to over 10x without user intervention.
January 26, 2020 by hgpu
