Re:Which Machines Do Computer Architects Admire? (2013)
by zozbot234 on Wednesday January 08, 2020 @12:31PM
(#21993646)
We need new programming models that make it easier to expose static parallelism to the compiler. Doing it all in plain old C/C++, or even in "managed" VM-based languages, cannot possibly work - and even conventional multi-threading is way too coarse-grained by comparison to what's most likely needed. Something based on a dataflow-oriented description of the code would probably work well, and be possible to integrate well enough with modern functional-like paradigms.
Re:Which Machines Do Computer Architects Admire? (2013)
by pdimitar on Wednesday January 08, 2020 @04:20PM
(#21995920)
I reached the same conclusion. We need a way to be able to explicitly say "this piece of code here is a standalone unit that is independent of everything else until we have to coalesce results of computation", or "this unit manages those other units and coalesces their results".
Erlang's OTP captures this pretty okay-ish with their concept of processes (preemptively scheduled green threads that get aggressively multiplexed on all CPU cores) but I feel we can go a little bit further than that and have some sort of shorter markers in the code, say `actor { ... }` or `supervises(a1, a2, a3) { ... }` or something.
Re:Which Machines Do Computer Architects Admire? (2013)
by Symmetry on Wednesday January 08, 2020 @04:29PM
(#21995998)
VISC seems at least potentially promising.
https://www.anandtech.com/show/10025/examining-soft-machines...
(I ended up running the OS group at Multiflow before bailing right before they hit the wall.)