On Tue, Jun 27, 2023 at 07:37:35PM +0200, Ralf Hemmecke wrote: > On 27.06.23 16:37, Waldek Hebisch wrote: > > > Actually, my question is whether it would be possible to clone a running > > > fricas session and continue with some computation in one clone with at the > > > same time with another computation in another clone? > > > > Well, that is what Unix fork is doing. > > OK, yes, that is a bit too heavy actually. What I like is to call a function > in parallel with different parameters and in the end combine the result.
Basically that is what Open MP is doing. But even with most efficient implementation this is heavy too. > That's useful for example with something like chinese remaindering. The > computations mod several primes are completely independent and should be > separated from each other, but it should be easily possible to combine the > results if all parallel computations have finished. Well, my student did such thing (in C), but that requires appropriate problem structure and scale. > > Well, I am not aware of a CAS that can work really in parallel fashion. > > Certainly a marketing document... > > https://www.wolfram.com/mathematica/compare-mathematica/files/ParallelProgrammingMapleMathematica.pdf > > but it looks as if there were some features in the direction that I > described above. Well, quickly looking at it I noticed mention of distributing code between nodes. So this looks like "generate code" approach, which can be quite good at large scale, but generating code by necessity takes time which may dominate at small scale. And Monte Carlo integration is classic example of "embarassingly parallel" code, which is easy to execute in parallel. Many special problems are "embarassingly parallel" but general problems solved by CAS usually are much harder to execute in parallel. So, yes, there is support for special cases, but not in general. -- Waldek Hebisch -- You received this message because you are subscribed to the Google Groups "FriCAS - computer algebra system" group. To unsubscribe from this group and stop receiving emails from it, send an email to fricas-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/fricas-devel/ZJtIrvVAHVg0U7bC%40fricas.math.uni.wroc.pl.
