[Oisf-users] Question about cpu-affinity

[Cooper F. Nelson]

Btw, the reason technologies like Intel's Hyperthreading and AMD's
“Clustered Multi-Threading” result in improved performance for typical
IO heavy multi-threaded workloads is entirely because of this effect. 
Since most of the threads are stalled most of the time due to cache
misses or waiting on IO, threads that are actually busy can make use of
all available execution units.  I suspect if you disabled HyperThreading
for your SEPTUN build  you would get similar or possibly better
performance, as Hyperthreading can actually impact performance if both
threads are attempting to make use of all available execution units. 

-Coop

On 3/4/2018 11:31 PM, Michał Purzyński wrote:
> That's
> nice, but those aren't even close to being saturated, as we're wasting
> cycles waiting for cache misses :/

-- 
Cooper Nelson
Network Security Analyst
UCSD ITS Security Team
cnelson at ucsd.edu x41042

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openinfosecfoundation.org/pipermail/oisf-users/attachments/20180306/6f84c489/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
URL: <http://lists.openinfosecfoundation.org/pipermail/oisf-users/attachments/20180306/6f84c489/attachment-0002.sig>