AI Chat AI Image Generator AI Video Text to Speech

Towards Ultra Rapid Restarts

02/18/2014

∙

by Shai Haim, et al.

∙

∙

We observe a trend regarding restart strategies used in SAT solvers. A few years ago, most state-of-the-art solvers restarted on average after a few thousands of backtracks. Currently, restarting after a dozen backtracks results in much better performance. The main reason for this trend is that heuristics and data structures have become more restart-friendly. We expect further continuation of this trend, so future SAT solvers will restart even more rapidly. Additionally, we present experimental results to support our observations.

Shai Haim
3 publications
Marijn Heule
8 publications

research

∙ 08/05/2019

Learned Clause Minimization in Parallel SAT Solvers

Learned clauses minimization (LCM) let to performance improvements of mo...

0 Marc Hartung, et al. ∙

research

∙ 06/08/2015

On SAT Models Enumeration in Itemset Mining

Frequent itemset mining is an essential part of data analysis and data m...

0 Said Jabbour, et al. ∙

research

∙ 05/11/2020

Designing New Phase Selection Heuristics

CDCL-based SAT solvers have transformed the field of automated reasoning...

0 Arijit Shaw, et al. ∙

research

∙ 10/10/2018

Incremental SAT Library Integration Using Abstract Stobjs

We describe an effort to soundly use off-the-shelf incremental SAT solve...

0 Sol Swords, et al. ∙

research

∙ 10/27/2021

An Experimental Study of Permanently Stored Learned Clauses

Modern CDCL SAT solvers learn clauses rapidly, and an important heuristi...

0 Sima Jamali, et al. ∙

research

∙ 09/02/2021

On Dedicated CDCL Strategies for PB Solvers

Current implementations of pseudo-Boolean (PB) solvers working on native...

0 Daniel Le Berre, et al. ∙

research

∙ 09/12/2013

Cache Performance Study of Portfolio-Based Parallel CDCL SAT Solvers

Parallel SAT solvers are becoming mainstream. Their performance has made...

0 Roberto Asín, et al. ∙