I’m increasingly convinced that future CPUs will be judged not by traditional FLOPS or iOPS metrics, but by aggregate transactional memory operations retired per second. All microarchitectural optimizations should be directed to this end.
-
-
(Hmm. This is true even of GHC's STM?)
-
Yes, Haskell STM sucks for hot spots
- 2 more replies
New conversation -
-
-
@TimSweeneyEpic check messages!Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.