Yes, it's very easy to make pause times low if you're willing to make allocation expensive - it's all trade-offs.
In multicore's case it's about keeping low pause times while also keeping allocation cheap _and_ maintaining throughput.
Re:Multicore OCaml: Feb 2020 update
by chrisseaton on Friday February 28, 2020 @03:36PM
(#22446094)
Why would allocation become more expensive in parallel? Surely you’re allocating in thread-local space? It’s like two machine instructions. Where does the extra overhead come from?
Both current multicore GCs do exactly that to keep allocation cheap but there are different design choices within the constraints we could have gone with.
To avoid frequent synchronisation we could have gone with a (potentially generational) non-moving collector for the minor GC which would still have preserved the C API, probably allowed for very low pauses but would have made allocation more expensive.
Re:Multicore OCaml: Feb 2020 update
by throwaway894345 on Friday February 28, 2020 @04:04PM
(#22446345)
If you want conventional shared memory parallelism, then your allocations can't assume thread-locality.
Re:Multicore OCaml: Feb 2020 update
by chrisseaton on Friday February 28, 2020 @04:05PM
(#22446354)
> If you want conventional shared memory parallelism, then your allocations can't assume thread-locality.
I don't really understand why not. Can you expand on it? Doesn't the JVM for example do thread-local allocations in a conventional shared-memory parallel environment?
Re:Multicore OCaml: Feb 2020 update
by throwaway894345 on Friday February 28, 2020 @04:56PM
(#22446788)
I was mistaken; wasn't thinking clearly as I responded. JVM can get away with this because it's generational/moving; bump allocator works in the young generation and then objects are subsequently moved. Generational/moving is tricky, especially with C APIs (if the GC moves an object that C has a pointer to, the C code dereferencing the pointer will find potentially garbage data) and I believe they make it difficult to get good STW times, but at this point I'm pretty well out of my depth.
I would expect that low pause times are easy enough (or relatively easy considering the difficult domain of GC optimization), but keeping the allocations cheap is (one of) the key constraints, no?