YewPar: Skeletons for Exact Combinatorial Search
- Submitting institution
-
Heriot-Watt University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 35213572
- Type
- E - Conference contribution
- DOI
-
10.1145/3332466.3374537
- Title of conference / published proceedings
- PPoPP '20: Proceedings of the 25th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming
- First page
- 292
- Volume
- -
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- February
- Year of publication
- 2020
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
-
- Citation count
- 1
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Exact combinatorial search is central to many applications and challenging to parallelise due to massive irregularity and search order dependence. Current implementations are for a single search. Originality: The first general-purpose scalable parallel framework for exact combinatorial search; allows non-experts to parallelise by reusing algorithmic skeletons; multiple architecture execution. Significance: premier parallelism conference; aims to establish combinatorial search as an HPC application domain (EPSRC CoDiMa EP/M022641). Rigour: A new formal model of parallel backtracking search uniformly characterises enumeration, decision and optimisation searches. Systematic performance analysis of 12 YewPar skeletons on standard instances of 7 search applications, scaling to 255 cores.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
