Also, you might consider using noop as an IO scheduler (or lack of one) since large SAN's (EVA8x00) handle command reordering at the controller level. This can make a difference and you can change it on the fly as well.
I also changed my nr_requests value to 256 up from 128 (512 was too much). These two things along with "group_by_prio" got me from 200MB's to > 500MB/s over 4G fibre. -C On Wed, Mar 17, 2010 at 11:18 AM, John Haxby <[email protected]> wrote: > > > On 16 March 2010 19:31, Troels Arvin <[email protected]> wrote: >> >> - Is there a way to adjust queue depths without unloading the >> qla2xxx kernel module? > > I believe so, eg: > > echo 128 > /sys/module/qla2xxx/parameters/ql2xmaxqdepth > > (In general the files under /sys/module/<module>/parameters that are > writeable are those than can be changed dynamically.) > > >> >> - If I set the queue depth to 128, is that >> - 128 in total for all devices handled by the qla2xxx >> module?, or >> - 128 for each HBA? (256 in total), or >> - each of the eight /dev/sdX devices (1024 in toal)?, or >> - each of the two /dev/mpath/... devices (256 in total)? >> This is important for me to know so that my server doesn't >> flood the storage system, affecting other servers connected >> to it. >> > > modinfo qla2xxx says (among other things): > > ql2xmaxqdepth:Maximum queue depth to report for target devices. > > I believe that each target corresponds to a /dev/sdX rather than each > multiply-connected LUN. > > jch > > _______________________________________________ > rhelv5-list mailing list > [email protected] > https://www.redhat.com/mailman/listinfo/rhelv5-list > > _______________________________________________ rhelv5-list mailing list [email protected] https://www.redhat.com/mailman/listinfo/rhelv5-list
