You are right, Thanks a lot Brian. rate(windows_logical_disk_read_write_latency_seconds_total[10s]) / (rate(windows_logical_disk_reads_total[10s]) + rate(windows_logical_disk_writes_total[10s]))
On Thursday, December 28, 2023 at 6:44:28 PM UTC+3 Brian Candler wrote: > Probably because you need to divide by the average number of operations > per second done in that time window. > > I can't give you the specifics for windows_exporter, but I can for Linux > node_exporter: > > https://brian-candler.medium.com/interpreting-prometheus-metrics-for-linux-disk-i-o-utilization-4db53dfedcfc > > The average latency for disk reads would be: > rate(node_disk_read_time_seconds_total[20s]) / > rate(node_disk_reads_completed_total[20s]) > > On Thursday 28 December 2023 at 13:53:51 UTC ضياء سامي wrote: > >> Hi >> >> I am trying to monitor Windows disk latency using windows_exporter. >> I tested disk performance using two commands below : >> C:\> diskspd.exe -c1G -d120 -b4K -o32 -t4 -w10 -L -D -Suw testfile.dat >> PS C:\> Get-Counter -ComputerName win-node -Counter '\LogicalDisk(*)\Avg. >> Disk sec/Read','\LogicalDisk(*)\Avg. Disk sec/Write' -SampleInterval 1 >> >> The output of both commands above shows average latency of 11 msec. >> >> While the prometheus query >> "rate(windows_logical_disk_read_write_latency_seconds_total[20s])" prints >> 125 >> >> Why windows_exporter showing different result ? >> > -- You received this message because you are subscribed to the Google Groups "Prometheus Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to prometheus-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/55ae07e9-e9db-47c5-bfc6-96d65d2fa777n%40googlegroups.com.
