https://community.splunk.com/t5/Monitoring-Splunk/Testing-Splunk-systems-with-IOzone-how-to-interpret-results/m-p/67902#M774 <P>Hmm, I don't know IOzone. I have used bonnie++</P> <P>Bonnie++ is available at <A href="http://www.coker.com.au/bonnie++">http://www.coker.com.au/bonnie++</A></P> <PRE><CODE>bonnie++ -d /volume -s &lt;2xmemory&gt; </CODE></PRE> <P><CODE>&lt;2xmemory&gt;</CODE> means twice the RAM on the system, and <CODE>/volume</CODE> is the path to local disk or NFS/SAN where the Splunk datastore will live. Splunk should not be running; in fact, Splunk doesn't need to be installed. Look for</P> <UL> <LI>sequential output rewrite; should be in the range 30,000-50,000 K/sec (affects indexing) </LI> <LI>sequential input block; should be &gt;100,000 K/sec (affects indexing)</LI> <LI>random seeks; should be &gt;1000/sec) (affects search)</LI> </UL> Sat, 03 Sep 2011 17:45:57 GMT lguinn2 2011-09-03T17:45:57Z https://community.splunk.com/t5/Monitoring-Splunk/Testing-Splunk-systems-with-IOzone-how-to-interpret-results/m-p/67901#M773 <P>Does anyone on here know how to interpret results from the IOzone disk benchmarking app to identify the results most relevant to Splunk?</P> <P>Categories:</P> <UL> <LI>Reader</LI> <LI>Writer</LI> <LI>Fread</LI> <LI>Fwrite</LI> <LI>Re-reader</LI> <LI>Re-writer</LI> <LI>Random Read</LI> <LI>Random Write</LI> <LI>Record Rewrite</LI> <LI>Re-fread</LI> <LI>Re-fwrite</LI> <LI>Backward Read</LI> <LI>Stride Read</LI> </UL> <P>The output data is in a table of "Record Sizes" from 4-16384, by "File Sizes" from 64 to 524288.</P> Wed, 31 Aug 2011 15:23:47 GMT https://community.splunk.com/t5/Monitoring-Splunk/Testing-Splunk-systems-with-IOzone-how-to-interpret-results/m-p/67901#M773 Jason 2011-08-31T15:23:47Z https://community.splunk.com/t5/Monitoring-Splunk/Testing-Splunk-systems-with-IOzone-how-to-interpret-results/m-p/67902#M774 <P>Hmm, I don't know IOzone. I have used bonnie++</P> <P>Bonnie++ is available at <A href="http://www.coker.com.au/bonnie++">http://www.coker.com.au/bonnie++</A></P> <PRE><CODE>bonnie++ -d /volume -s &lt;2xmemory&gt; </CODE></PRE> <P><CODE>&lt;2xmemory&gt;</CODE> means twice the RAM on the system, and <CODE>/volume</CODE> is the path to local disk or NFS/SAN where the Splunk datastore will live. Splunk should not be running; in fact, Splunk doesn't need to be installed. Look for</P> <UL> <LI>sequential output rewrite; should be in the range 30,000-50,000 K/sec (affects indexing) </LI> <LI>sequential input block; should be &gt;100,000 K/sec (affects indexing)</LI> <LI>random seeks; should be &gt;1000/sec) (affects search)</LI> </UL> Sat, 03 Sep 2011 17:45:57 GMT https://community.splunk.com/t5/Monitoring-Splunk/Testing-Splunk-systems-with-IOzone-how-to-interpret-results/m-p/67902#M774 lguinn2 2011-09-03T17:45:57Z