On Thu, 28 Dec 2017 17:16:40 -0500
Paul Davis <paul@linuxaudiosystems.com> wrote:
>On Thu, Dec 28, 2017 at 2:11 PM, Jeremy Henty <onepoint@starurchin.org>
>wrote:
>
>>
>> Will Godfrey wrote:
>>
>> > If I've understood that correctly you can also ensure that they are
>> > also on the same socket, which apparently improves memory access.
>>
>> I think this is what is meant by NUMA (Non-Uniform Memory access).
>>
>
>AFAIK, NUMA is dead for everything except a few research systems.
>
>Parallel/multi-processor systems these days are all "symmetric" (all
>processors have symmetrical access to all memory).
>
>NUMA is really, really, really hard to get right. Why? Cache invalidation.
>Several companies, organizations, etc. have tried. Last time I looked (and
>it has been a while, but I was quite involved with this stuff in the mid
>1990s), everybody failed.
>
This is proving to be quite fascinating, so thanks everyone for jumping on
board and contributing.
I won't say the mists have cleared, but they certainly seem a lot thinner :)
-- Will J Godfrey http://www.musically.me.uk Say you have a poem and I have a tune. Exchange them and we can both have a poem, a tune, and a song. _______________________________________________ Linux-audio-user mailing list Linux-audio-user@lists.linuxaudio.org https://lists.linuxaudio.org/listinfo/linux-audio-userReceived on Sat Dec 30 00:15:01 2017
This archive was generated by hypermail 2.1.8 : Sat Dec 30 2017 - 00:15:01 EET