Performance Impact of Precision Reduction In Sparse Linear Systems Solvers

Zounon, Mawussi and Higham, Nicholas J. and Lucas, Craig and Tisseur, Françoise (2020) Performance Impact of Precision Reduction In Sparse Linear Systems Solvers. [MIMS Preprint]

This is the latest version of this item.

[img] Text
paper.pdf

Download (631kB)

Abstract

It is well established that reduced precision arithmetic can be exploited to accelerate the solution of dense linear systems. Typical examples are mixed precision algorithms that reduce the execution time and the energy consumption of parallel solvers for dense linear systems by factorizing a matrix at a precision lower than the working precision. Much less is known about the efficiency of reduced precision in parallel solvers for sparse linear systems, and existing work focuses on single core experiments. We evaluate the benefits of using single precision arithmetic in solving a double precision sparse linear system using multiple cores. We consider both direct methods and iterative methods and we focus on using single precision for the key components of LU factorization and matrix–vector products. Our results show that the anticipated speedup of 2 over a double precision LU factorization is obtained only for the very largest of our test problems. We point out two key factors underlying the poor speedup. First, we find that single precision sparse LU factorization is prone to a severe loss of performance due to the intrusion of subnormal numbers. We identify a mechanism that allows cascading fill-ins to generate subnormal numbers and show that automatically flushing subnormals to zero avoids the performance penalties. The second factor is the lack of parallelism in the analysis and reordering phases of the solvers and the absence of floating-point arithmetic in these phases. For iterative solvers, we find that for the majority of the matrices computing or applying incomplete factorization preconditioners in single precision provides at best modest performance benefits compared with the use of double precision. We also find that using single precision for the matrix–vector product kernels provides an average speedup of 1.5 over double precision kernels. In both cases some form of refinement is needed to raise the single precision results to double precision accuracy, which will reduce performance gains.

Item Type: MIMS Preprint
Uncontrolled Keywords: Mixed precision algorithms, iterative refinement, sparse linear algebra, subnormal numbers
Subjects: MSC 2010, the AMS's Mathematics Subject Classification > 15 Linear and multilinear algebra; matrix theory
MSC 2010, the AMS's Mathematics Subject Classification > 65 Numerical analysis
MSC 2010, the AMS's Mathematics Subject Classification > 68 Computer science
Divisions: Manchester Institute for the Mathematical Sciences
Depositing User: PhD Mawussi Zounon
Date Deposited: 04 Jun 2021 09:19
Last Modified: 04 Jun 2021 16:21
URI: http://eprints.maths.manchester.ac.uk/id/eprint/2811

Available Versions of this Item

Actions (login required)

View Item View Item