2017/09/07 by John C. McCabe-Dansted, Mark Reynolds
Computer Science · #cs.LO #cs.PF
paper · pdf · doi:10.4204/eptcs.256.12
published as EPTCS 256, 2017, pp. 166-179 · In Proceedings GandALF 2017, arXiv:1709.01761
arxiv created 2017/09/07 · arxiv updated 2017/09/08
For many applications, we are unable to take full advantage of the potential massive parallelisation offered by supercomputers or cloud computing because it is too hard to work out how to divide up the computation task between processors in such a way to minimise the need for communication. However, a recently developed branch-independent tableaux for the common LTL temporal logic should intuitively be easy to parallelise as each branch can be developed independently. Here we describe a simple technique for partitioning such a tableau such that each partition can be processed independently without need for interprocess communication. We investigate the extent to which this technique improves the performance of the LTL tableau on standard benchmarks and random formulas.