On Tue, May 23, 2017 at 5:34 PM, Jeff Furlong <jeff.furlong@xxxxxxx> wrote: > Generally speaking, it is best to preconditon with the block size you intend to test. There are some exceptions but they are device dependent and vary from one SSD to another. Depending upon your quality of service test (are you measuring steady state IOPs or latency or something else?), you may notice these differences in your 8k results. > > Regards, > Jeff > > -----Original Message----- > From: fio-owner@xxxxxxxxxxxxxxx [mailto:fio-owner@xxxxxxxxxxxxxxx] On Behalf Of Matthew Eaton > Sent: Monday, May 22, 2017 3:32 PM > To: fio@xxxxxxxxxxxxxxx > Subject: SSD Test Recommendation > > Hi guys, > > I want to run a quality of service test on SSDs with bs=4k and bs=8k but I need some help. > I'm wondering if it's okay to test both 4k then 8k like this: > > purge > precondition with 128k write > run 4k read test > run 8k read test > precondition to steady state with 4k write run 4k write test run 8k write test > > Will the 8k test results be invalid due to the 4k write history? > > Thanks! > -- > To unsubscribe from this list: send the line "unsubscribe fio" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html > Western Digital Corporation (and its subsidiaries) E-mail Confidentiality Notice & Disclaimer: > > This e-mail and any files transmitted with it may contain confidential or legally privileged information of WDC and/or its affiliates, and are intended solely for the use of the individual or entity to which they are addressed. If you are not the intended recipient, any disclosure, copying, distribution or any action taken or omitted to be taken in reliance on it, is prohibited. If you have received this e-mail in error, please notify the sender immediately and delete the e-mail in its entirety from your system. Hi Jeff, Thanks for your insight. I was leaning the same way but figured I would consult the experts first. :) -- To unsubscribe from this list: send the line "unsubscribe fio" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
