Latest Posts



Translate

Total Pageviews

Friday, 15 November 2013

vSCSIStats

The vscsistats tool provides virtual disk latency statistics for VMFS and NFS datastores. With microsecond precision, it displays data in a histogram form for trend analysis. This precise data collection is superior to means, averages, and standard deviations when you are trying to determine the sequentiality, I/O size, and block distribution of your datastore’s workload.
Use vscsiStats to gather storage performance data
Note: vscsistats is located in the /usr/sbin directory
  1. Determine the world number for your virtual machine:
    From an SSH session run: vscsiStats -lRecord the world ID for the virtual machine you would like to monitor. 
  2. Start the collection process:
    vscsiStats -s -w worldID
  3. This collection process will continue for 30 minutes unless stopped and the statistics are cleared using this command:
    vscsiStats -x -w worldID
  4. To extend the collection process by 30 minutes, rerun this command:
    vscsiStats -s -w worldID 
  5. To only reset the counters to zero while collecting statistics, execute:
    vscsiStats -r 
Display histograms
Histogram data may be outputted as a chart or a comma separated file for the following metrics: seekDistance, ioLength, outstandingIOs, latency, interarrival, or all statistics.
Command Arguments
  • seekDistance is the distance in logical block numbers (LBN) that the disk head must travel to read or write a block. If a concentration of your seek distance is very small (less than 1), then the data is sequential in nature. If the seek distance is varied, your level of randomization may be proportional to this distance traveled.
  • interarrival is the amount of time in microseconds between virtual machine disk commands.
  • latency is the time of the I/O trip.
  • ioLength is the size of the I/O. This is useful when you are trying to determine how to layout your disks or how to optimize the performance of the guest O/S and applications running on the virtual machines.
  • outstandingIOs will give you an idea of any queuing that is occurring.
Seek Distance as a chart
  • vscsiStats -p seekDistance
Latency as a csv
  • vscsiStats -p latency -c ./chart.out
Display all statistics
  • vscsiStats -p all