4 ms·

Thank you for the detailed reply, this is the first cogent answer I have read that explains why GPU solvers are not as widely used in SAT. The argument that we

Thank you for the detailed reply, this is the first cogent answer I have read that explains why GPU solvers are not as widely used in SAT. The argument that we should focus on getting sequential solvers to run fast, then parallelizing that in hopes that it can be reused on each core does make sense, although I can't help but wonder if there isn't a smarter way to leverage parallelism than simply running CDCL in parallel on n-independent cores. I guess if I were to summarize your argument, (current) solvers do not use GPUs because there are not many algorithms that can leverage the additional compute (and maybe there is some theoretical barrier?). On the other hand, the continuous optimization folks had neural networks for more than half a century before they figured out how to make them run fast, and were also ridiculed for pursing what was widely believed to be a dead-end. So I would be careful about falling for the "smart people tried really hard for a while" argument.

calf 3y ago
Yeah if there indeed is some kind of theoretical reason that makes NP-complete problems hard to parallelize, that would be interesting to know about and understand. Maybe somebody has studied this problem. I think that unlike the story with neural nets, there wasn't some faction trying to say that parallelizing SAT is a bad idea. In fact the state-of-the-art sequential solvers use techniques to exploit properties of CPU caches to get their speedup breakthrough; that in itself was an important lesson within the research community to pay close attention to hardware capabilities. My impression is that ever since that success within their discipline, SAT specialists became interested not just in algorithms/formal theory but also hardware and implementation, and in today's case that would include having a close look at GPUs/TPUs for the possibility of new algorithms.

calf 3y ago

Yeah if there indeed is some kind of theoretical reason that makes NP-complete problems hard to parallelize, that would be interesting to know about and understand. Maybe somebody has studied this problem. I think that unlike the story with neural nets, there wasn't some faction trying to say that parallelizing SAT is a bad idea. In fact the state-of-the-art sequential solvers use techniques to exploit properties of CPU caches to get their speedup breakthrough; that in itself was an important lesson within the research community to pay close attention to hardware capabilities. My impression is that ever since that success within their discipline, SAT specialists became interested not just in algorithms/formal theory but also hardware and implementation, and in today's case that would include having a close look at GPUs/TPUs for the possibility of new algorithms.