Well, The correct answer to all interview questions is valid for this topic as well.
IT DEPENDS :)
If we assume a box with 8 Core CPUs with 8MB of L3 Cache (as can be seen in the picture below) each core gets 1MB of cache. When Hyperthreading is enabled you have 16 cores (logical processors) and a core has only 0.5 MB of cache left.
Well It definately cuts Cache per Core but You have to test it with your workload.
It all depends on how your system behaves with HT enabled or not. :)
I found this article useful, about L3 Cache
http://www.extremetech.com/computing/55662-top-tip-difference-between-l2-and-l3-cache
and thanks http://www.brentozar.com/ for the original explanation.
Cheers!!
IT DEPENDS :)
If we assume a box with 8 Core CPUs with 8MB of L3 Cache (as can be seen in the picture below) each core gets 1MB of cache. When Hyperthreading is enabled you have 16 cores (logical processors) and a core has only 0.5 MB of cache left.
Well It definately cuts Cache per Core but You have to test it with your workload.
It all depends on how your system behaves with HT enabled or not. :)
I found this article useful, about L3 Cache
http://www.extremetech.com/computing/55662-top-tip-difference-between-l2-and-l3-cache
and thanks http://www.brentozar.com/ for the original explanation.
Cheers!!
