tis 2007-03-20 klockan 21:09 -0300 skrev Michel Santos: > I do compare the incoming http traffic to the outgoing. Higher the > difference better my cache performance right. The better hit ratio you have. But tells nothing about the performance. An overloaded disk can be significantly slower than a fast Internet connection. > that is certainly an interesting point. IO Bound I guess can be fight by > faster and cpu independent disks and subsystem (scsi) and then using > polling on for example em (intel pro) nics which seem to "produce" less > interrupts. None of these helps in speeding up the rotation and seek time of a disk.. It's physical limitaions of things moving around. > Also setting vfs.write_behind and vfs.vmiodirenable may give important > improvement on some hardware together with vfs.read_max. Not familiar with FreeBSD terminology. > All this does not cut ufs's bottleneck but helps a lot. So sure diskd is > the preferred cache_dir on FreeBSD. I would say aufs is the preferred cache_dir on FreeBSD, Linux and Solaris these days. aufs requires POSIX kernel threads, which is available even on FreeBSD these days. > But again, not on low traffic machines > where I can not find any difference. IMO so long as your machine does not > handle more than 2mb/s it does not matter what you do FreeBSD does it > well either way - supposed you have good hardware. With only 2mb/s you are unlikely to reach even 30 req/s, so yes.. and this not even needing good hardware just not too crappy hardware. Regards Henrik
Attachment:
signature.asc
Description: Detta =?ISO-8859-1?Q?=E4r?= en digitalt signerad meddelandedel
