It may not be relevant to anyone, but the article below seemed interesting as that, my "guess" is that performance can be optimized by increasing the performance of the disk "read" (ahead) cache and to a certain extent the read block size.
However, my fear is that depending on the drivers, my fear is that tweaks in this are may not lead to any significant performance increases. I am sure there is no silver bullet other than having several independent spindles associated with a group of volumes as described by Richard and having a single rrd process dedicated to each volume. Regards, Larry Adams The Cacti Group http://docsrv.sco.com:507/en/PERFORM/increase_buffer_cache.html -- Unsubscribe mailto:[EMAIL PROTECTED] Help mailto:[EMAIL PROTECTED] Archive http://lists.ee.ethz.ch/rrd-developers WebAdmin http://lists.ee.ethz.ch/lsg2.cgi
