we mirror lvs across 2 storage systems (local, remote). remote storage has less io throughput than local. with the default lv scheduling policy lvm will write in parallel and read form the "first" disk until its 100% busy and then use the "second" mirrored disk. so for writing, the remote disk throughput limits the overall mirror IO performance. no luck there.
for reading I like to make sure that the first disk chosen is a local one. iostat shows one of the disks of a mirror getting far more reads than the other one.
does anyone know how the "first" disk in this algorithm is chosen? from iostat I suspect that the lower hdisk# will be used as first disk and this disk will do the majority of read IO.
i would like to control this to make a disk from local storage always the first one chosen for reads.
Any input welcome
Joachim
Tags
Use the search field to
find all types of content in My developerWorks with that tag.
Use the slider bar to see more or fewer tags.
Popular tags shows the top tags for this particular type of content or application that you're viewing.
My tags shows your tags for this particular type of content or application that
you're viewing.