Values of factors considered when isolating Storage Performance issues- Please guide
Been working as SAN consultant, I am always curious to know the various parameters which a Sr. SAN Consultant/Oncall person study/review when there are any issues/alerts reported on SAN storage to them critical with less time.
I am curious to take up the responsiblities from oncall perceptive and seeking guidance on the major factor one consideres while isolating an issue related to SAN Performace degrade.
what are specific values (threshold, eg 15 ios/sec, 30% cache writing pending etc ) which are the watermarks for those specific paramteres one reviews from monitoring tool like HTm, SMC, Perf Advisor.
( Eg. Read cache Pending, write cache pending, IOS/sec, back end and front end adapters ).
And are these values same across all vendor arrays.
Software/Hardware used:
Software/Hardware used:
ASKED:
January 2, 2013 11:28 PM
UPDATED:
January 3, 2013 1:37 PM
Answer Wiki:
I typically look at the response time and the service time to start with as these are going to tell me how long the array is taking to service the request. If the service time is high and the number of IOPs is low I know that there's something going wrong in the array somewhere.
If the response time and service time are low (single digit milliseconds) then it's a server problem and I can start looking there.
All Answer Wiki Contributors:
Denny Cherry 
64,520 pts. ,
Michael Tidmarsh 11,380 pts. ,
StorageSAN 5 pts.
64,520 pts. ,
Michael Tidmarsh 11,380 pts. ,
StorageSAN 5 pts. To see all answers submitted to the Answer Wiki: View Answer History.
RSS - Discussions Help
Discuss This Question:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
