Re: tunning disk cache for pgsql?



On Dec 28, 2010, at 7:59 AM, Nicolas Haller wrote:
I use a new box with 4GB RAM as a pgsql server. In pgsql, you can
set the effective_cache_size to indicate the memory available to cache
disk I/O.
As "recommended", my box use 1300MB to shared buffers (IPC shared memory)
and 2700 Mo to disk cache.

That's probably not a great mix unless your workload is very read-heavy. Writes will push data through shared buffers back into the OS, which will also try to cache it, so you'll end up with double-buffering.

If I look memory usage in top, it say:
Mem: 1154M Active, 1911M Inact, 601M Wired, 112M Cache, 417M Buf, 148M Free

The Cache reported by top in FreeBSD isn't filesystem cache; it's a cache for some internal stuff. Buf are filesystem buffers, but they're not the only mechanism for the OS to cache data. Most data is actually cached via active and inactive pages.
--
Jim C. Nasby, Database Architect jim@xxxxxxxxx
512.569.9461 (cell) http://jim.nasby.net


_______________________________________________
freebsd-performance@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-performance
To unsubscribe, send any mail to "freebsd-performance-unsubscribe@xxxxxxxxxxx"



Relevant Pages

  • 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)
  • Re: Problem: Creating a raw binary string
    ... > While its true that a 64-bit cpu will move twice the data per instruction it ... > Memory bus width plays an important role here and unless it too is widened / ... You are forgetting the two levels of cache in the processor. ... The memory chips are addressed in Row col fashion. ...
    (alt.comp.lang.borland-delphi)
  • Re: Is Greenspun enough?
    ... Most OSes memory map executables directly from the file system so code doesn't pollute the file cache or swap space. ...
    (comp.lang.lisp)