We use huge pages here as a standard. How do people test huge pages to
isolate variables so you can measure the improvement from huge pages?
On Sat, Jul 15, 2017 at 3:09 AM, Stefan Koehler <contact@xxxxxxxx> wrote:
Hey,
the reason behind your performance improvement is a better TLB usage - you
don't need to work through the page table so often (translation physical
address from a virtual address) if you are using HugePages.
VMware explained how TLB is used with their software here (page 3&4):
https://www.vmware.com/content/dam/digitalmarketing/
vmware/en/pdf/techpaper/vsp_41_perf_memory_resource-
management-white-paper.pdf
@ Scott: I like a sentence that Mark Bobak said: "If you’re not using
hugepages, you’re doing it wrong!"
Best Regards
Stefan Koehler
Independent Oracle performance consultant and researcher
Website: http://www.soocs.de
Twitter: @OracleSK
Andrew Kerber <andrew.kerber@xxxxxxxxx> hat am 15. Juli 2017 um 01:21geschrieben:
or better) performance improvement when using hugepages on Virtual Machines.
we have had good luck with them Our observation is a substantial (10%
--
Andrew W. Kerber
//www.freelists.org/webpage/oracle-l