Closing the Performance Gap between Doubles and Rationals for Octagons

Chawdhary, Aziem and King, Andy (2018) Closing the Performance Gap between Doubles and Rationals for Octagons. In: Podelski, Andreas, ed. International Static Analysis Symposium. Lecture Notes in Computer Science, 11002 . Springer, pp. 187-204. ISBN 978-3-319-99724-7. E-ISBN 978-3-319-99725-4. (doi:https://doi.org/10.1007/978-3-319-99725-4_13) (Full text available)

Abstract

Octagons have enduring appeal because their domain opera- tions are simple, readily mapping to for-loops which apply max, min and sum to the entries of a Difference Bound Matrix (DBM). In the quest for efficiency, arithmetic is often realised with double-precision floating- point, albeit at the cost of the certainty provided by arbitrary-precision rationals. In this paper we show how Compact DBMs (CoDBMs), which have recently been proposed as a memory refinement for DBMs, enable arithmetic calculation to be short-circuited in various domain operations. We also show how comparisons can be avoided by changing the tables which underpin CoDBMs. From the perspective of implementation, the optimisations are attractive because they too are conceptually simple, following the ethos of Octagons. Yet they can halve the running time on rationals, putting CoDBMs on rationals on a par with DBMs on doubles.

Item Type: Book section
Divisions: Faculties > Sciences > School of Computing > Programming Languages and Systems Group
Depositing User: Andy King
Date Deposited: 03 Dec 2018 10:55 UTC
Last Modified: 04 Dec 2018 11:36 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/70572 (The current URI for this page, for reference purposes)
  • Depositors only (login required):

Downloads

Downloads per month over past year