The setup of a Nutanix appliance is very simple and straight forward. After it’s been set up, it’s also very simple and straight forward to run a quick performance diagnostics to ensure everything is running as expected. To do this you can use diagnostics.py from the shell of one of the Nutanix Controller VM’s (CVM). Below is the command I use when doing this.
Here is the command I use to run diagnostics.py from the Nutanix user home directory on my CVM’s.
diagnostics/diagnostics.py – -display_latency_stats – -run_iperf run ; diagnostics/diagnostics.py cleanup
Unlike my 4 Son’s, this command also cleans up after itself. Note: before each command you need to make sure you have two dashes or minus signs, i.e. – -. Copy and paste from this page may not pick that up.
This quick test will give you results in terms of IOPS, throughput (MBPS) and also iPerf network performance between each CVM. For the random read and write tests it uses 8 outstanding IO operations and 4KB IO size. For throughput tests it uses 1MB IO Size and provides throughput in terms of MBPS.
Simple test, simple results, not application realistic. But it’s a start to see if you at last have things set up well. This is an easy way to highlight any issues with network configuration. I also recommend that you run a Nutanix Cluster Check (NCC) after deployment to ensure everything is as expected. To do so simply run ncc health_checks run_all from the CVM. You can also view the health status of your cluster from PRISM in the Cluster Health User Interface.
This post first appeared on the Long White Virtual Clouds blog at longwhiteclouds.com. By Michael Webster +. Copyright © 2012 – 2015 – IT Solutions 2000 Ltd and Michael Webster +. All rights reserved. Not to be reproduced for commercial purposes without written permission.