I recently had a run-in with a 3x performance diff for @rustlang on the new @amd Threadripper with multi threadded applications. It seems to be related to the layout, I wonder if anyone is interested in me digging deeper into this?
-
-
That’s changed however with Zen 2, where it gest more complicated to the point that I think NUMA isn’t enough to express it any more. The 3000 chips are a single NUMA core, at least from the point of view of the M - Memory, they moved the controller from the CCD to the IO Die
-
That’s not entirely correct it seems, just reading up on the architecture, even so it’s on the IO die it’s still assigned a to a compute unit so there is still affinity.
End of conversation
New conversation -
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.

