One of the volumes for one of our (Ubuntu 16.04) Google Virtual Virtual Machines is used at 100% disk usage almost all the time - here is an example of 10 seconds collected randomly from the system:
iostat -x 10
Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await r_await w_await svctm %util
sdd 0.60 17.20 5450.50 2468.00 148923.60 25490.00 44.05 11.81 1.49 1.13 2.29 0.13 99.60
This is currently a 2.5T solid state SSD.
My understanding is that I cannot get better performance by adding virtual "spindles" and then distributing the workload through them.
This is a database volume, so I cannot use volatile SSD either.
I currently have XFS with these settings:
type xfs (rw,noatime,nodiratime,attr2,nobarrier,inode64,noquota)
Any suggestions?
source
share