Re: page fault & degaradation performance



* perryh@xxxxxxxxxxxxxx (perryh@xxxxxxxxxxxxxx) wrote:

how to reduce the number of page faults to upgrade program or OS
performance?

Install more memory.

I can cause paging on my hugely overendowed RELENG_6 system (8GB, ~2GB
active) just by reading some big files and churning through cache. It's
nice seeing syslogd go into pfault while there's 5G of inactive memory
and 300MB cache just waiting to be recycled.

There would seem to be some room for tuning on at least some systems.

--
Thomas 'Freaky' Hurst
http://hur.st/
_______________________________________________
freebsd-hackers@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to "freebsd-hackers-unsubscribe@xxxxxxxxxxx"



Relevant Pages

  • Re: MFC executable is 100% slower on faster machine
    ... Cache replacement algorithm ... Memory architecture ... Sounds like paging to me. ...
    (microsoft.public.vc.mfc)
  • Re: MFC executable is 100% slower on faster machine
    ... Cache replacement algorithm ... Memory architecture ... Sounds like paging to me. ... The slower machine has twice the memory, ...
    (microsoft.public.vc.mfc)
  • Re: Cached memory never gets released
    ... Stock linux 2.4.26 kernel. ... Due to flash bug 3M of memory gets lost due to font memory getting lost ... The output of "free" cache number steadily grows. ... longer to exhaust all of system memory with the cache. ...
    (Linux-Kernel)
  • Re: UMA cache back pressure
    ... done in memory allocation last years improved situation. ... limits on cache sizes -- they are self-tuned, ... Even on my workload with ZFS creating strong memory pressure I still have mbuf* zones buckets almost maxed out. ...
    (freebsd-current)
  • Re: UMA cache back pressure
    ... done in memory allocation last years improved situation. ... limits on cache sizes -- they are self-tuned, ... Even on my workload with ZFS creating strong memory pressure I still have mbuf* zones buckets almost maxed out. ...
    (freebsd-hackers)