Hi, Vincent > I think the better solution is to augment the documentation with a > detailed discussion of the file operation ioengines because they work so > differently from other fio ioengines. > Because of how unusual these ioengines are, there are many ways the > ioengines can be misused and their results misinterpreted. Making a > change like this reduces the possible confusion by only a small amount. > Having a detailed guide to these ioengines in the documentation will > more comprehensively reduce confusion and misuse of these ioengines. I agree with you. File operation engines are different from IO engines. Making detailed explanation in document is necessary. Meanwhile, it is still possible to make runtime job definition and log output more friendly to user with only small modification to code. This patch corrects 'rw' field in job file and log, next I plan to correct IOPS, BW fields in log. With several steps of "not-big" modification gradually, it is safe to make FIO fit for file operation measurement. I have sent the Patch V2 which includes runtime modification and doc update. In our practice, file operation performance is often measured. It is as same importance as IO performance. It will be efficient if we can utilize FIO's powerful framework and various statistics for file operation measurement. Best Regards Friendy -----Original Message----- From: Vincent Fu <vincentfu@xxxxxxxxx> Sent: Wednesday, November 30, 2022 1:42 AM To: Su, Friendy <Friendy.Su@xxxxxxxx>; fio <fio@xxxxxxxxxxxxxxx> Cc: Aoyama, Wataru (SGC) <Wataru.Aoyama@xxxxxxxx>; Kobayashi, Kento (SGC) <Kento.A.Kobayashi@xxxxxxxx> Subject: Re: [PATCH v1] fio: Append "filecreate/filestat/filedelete" to 'readwrite=' On 11/21/22 05:36, Friendy.Su@xxxxxxxx wrote: >> I think having this option will be confusing and lead to users trying to >> run a job with --ioengine=psync --rw=filecreate. > >> However, I understand the confusion that arises when the file operation >> ioengines are used. Might there be some other way to help? Perhaps the >> ioengine could emit some sort of message? Or the documentation for the >> file operation ioengines could be enhanced? > > I created this patch because somebody sent me email to ask why rw string shows 'read' when he measured by filecreate engine. And more, users felt strange what does IOPS/BW for file operation mean? > I think we can do option check at beginning. '--ioengine=filecreate' and '--rw=filecreate' must be paired. If not paired, emit error message and exit. How do you think? > > Though FIO is designed for IO measurement originally, file operation performance is also regularly evaluated for file system. It is helpful if expand FIO to adapt to file operation benchmark. > > Friendy > I think the better solution is to augment the documentation with a detailed discussion of the file operation ioengines because they work so differently from other fio ioengines. Because of how unusual these ioengines are, there are many ways the ioengines can be misused and their results misinterpreted. Making a change like this reduces the possible confusion by only a small amount. Having a detailed guide to these ioengines in the documentation will more comprehensively reduce confusion and misuse of these ioengines. Vincent
