On Wed, 2013-06-12 at 13:02 +0200, Carlos E. R. wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 2013-06-12 11:04, Roger Oberholtzer wrote:
Obviously a brute-force approach that is really only possible on my system as it does not seem to mess up general usage. The rate of 60 seconds is arbitrary. But each time this is run, the cache has grown to almost 3 GB.
As you generate 25E6 (roughly) per second, after 60" you gave generated 1.5GB (not 1.5GiB). The delay you had was 5". In that time the disk system can write 0.5GB at 100MB/s - so that delay is not flushing the entire cache. Not enough time at all.
I know. It is like the cache grows, and a small part is typically flushed. And then, occasionally, a bit more gets flushed that takes more time. I don't really have a problem with how it flushes things, as long as two things could be met: * the cache does not grow endlessly to full RAM * the flush, when it happens, does not take too long. A suggestion on the XFS list is that I may need to tweak some variables, as suggested by Lars Müller and others earlier in the thread. I just have not figured them out to know what to change. I am working on that... Yours sincerely, Roger Oberholtzer Ramböll RST / Systems Office: Int +46 10-615 60 20 Mobile: Int +46 70-815 1696 roger.oberholtzer@ramboll.se ________________________________________ Ramböll Sverige AB Krukmakargatan 21 P.O. Box 17009 SE-104 62 Stockholm, Sweden www.rambollrst.se -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org