Tagged as Erlang, Lecture, Programming Languages
Written on 2007-08-17 12:27:58
So, I apparently asked a good question
on Patrick Logan's blog
. And he gave me a good answer:
"I can see someone making the argument in some domain I don't usually work in that shared memory threads are better than shared nothing message passing for performance reasons. Some hard-real time scenario, some huge number crunching scenario, etc. where every byte and every cycle has to count in the extreme. But the irony then is that STM seems far more suited to a language like Haskell, which is also unlikely to be suited for these performance scenarios.
My only fear is that for the masses including myself, we need *simple* mechanisms, and the fewer of those, the better. Shared nothing messages seem to scale up, out, and down. STM seems more complicated, and an incomplete solution, and only a solution for shared memory."
Indeed. Aside from Andre Pang's Erlang and Concurrency talk
, I've watched Adam Welc's Google Tech Talk on Transactional Memory
and am planning on listening to Simon Peyton-Jones OSCON talk on STM
later today. Now, whether I'll end up listening to the other two
Google Tech Talks on Transaction Memory I don't know. I have, as mentioned, read Sweeney's POPL slides
So far, I end up thinking two things hearing these talks.
1) This is an ugly hack.
2) Overhead, overhead, overhead.
At any rate, It's nice to be involved in a high-level nerd conversation with very smart people. That's always fun.