November 17, 2011 at 3:20 am
Im doing some benchmarking of SAN's to judge if they are usable in a future project.
Cache size on the SAN is 256GB or there abouts. I have created files of 800GB in size on a number of Luns to simulate a database of considreable size and also to saturate the cache.
One thing im unsure of though is test duration. Collegues say if you run the test once to fill the cache and then run the actual test,30-60 seconds is plenty of time to saturate the cache and get accurate results.
Is this a valid statement? Considering the number of tests i need to perform, 30 seconds is appealing, but if its not accurate im happy to change it.
November 18, 2011 at 11:27 am
You may want to take a look at document next: http://msdn.microsoft.com/en-us/library/cc966412.aspx ... it describes Microsoft Best Practices for testing I/O subsystem capacity and performance.
_____________________________________
Pablo (Paul) Berzukov
Author of Understanding Database Administration available at Amazon and other bookstores.
Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply