Re: numbers don't lie ...

On 09/20/06 07:50, Oliver Fromme wrote:
Dmitry Morozovsky wrote:
> Oliver Fromme wrote:
> > Because buildworld is I/O-bound on systems with sufficiently
> > fast processors.
> > > > Try putting the contents of /usr/src into a RAM disk and
> > repeat the benchmark. The numbers might look a little
> > different then. Of course, you should have sufficient RAM
> > in the machines -- If they're going to swap to the disks,
> > your benchmark won't be happy.
> > > > I think putting /usr/obj onto a RAM disk is _not_ necessary
> > because of soft-updates, so the processes shouldn't block
> > on writes.
> > My experiments show that if you have enough memory to host radmdrive for > /usr/src you'd better leave it for caching - there were no statistically
> meaningful performance difference, at least on machines with 1G+ RAM.

That might only be true if you have enough RAM to keep
_all_ buildworld files (src, obj, toolchain) in the cache
_and_ you pre-read all of /usr/src before actually starting
the buildworld, so it is in the cache. If you don't have
that much RAM, but enough to store /usr/src, then using
a RAM disk for it is a win.

Reading /usr/src from a physical disk certainly requires
quite some I/O that takes more than zero time.


But, in order to populate the ram disk, you must read /usr/src also from something, and that also takes time, which you should include in the full scope.

Eric



--
------------------------------------------------------------------------
Eric Anderson Sr. Systems Administrator Centaur Technology
Anything that works is better than anything that doesn't.
------------------------------------------------------------------------
_______________________________________________
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: numbers dont lie ...
    ... OF> fast processors. ... you should have sufficient RAM ... OF> your benchmark won't be happy. ... at least on machines with 1G+ RAM. ...
    (freebsd-hackers)
  • Re: numbers dont lie ...
    ... fast processors. ... you should have sufficient RAM ... your benchmark won't be happy. ... a RAM disk for it is a win. ...
    (freebsd-hackers)
  • Re: Itanium Madison blasts Sun, IBM in encryption specs!
    ... Public assertions of SPEC unpublished benchmark performance figures ... RAM on the V210 is 1/2 that of the V240: ... Now that does not include a credit for the 2GB of RAM that needs to be ... The cost for ZWS is $1700 for an up to two CPU system. ...
    (comp.os.vms)
  • Re: cache limit
    ... > That was because they wanted the non-streaming files to be left in the cache. ... >>RAM are going to make a difference. ... [root@hera ifilipau]# hdparm -t /dev/hda ... To be bsolutely fair in this unfair benchmark I have run test only ...
    (Linux-Kernel)
  • SP2 slowed my virtual machine 50%
    ... I would benchmark my new VM compared my old VM. ... I downloaded the SP2 standalone installer on ... with 640MB RAM. ...
    (microsoft.public.mac.virtualpc)