marty wrote: > Faster to do what? OS A is faster at handling character-at-a-time data, > OS B is faster at handling block-at-a-time data.
I could be wrong, but given the big difference in read performance, I suspect bonny maybe getting confused about filesystem block sizes. I'd be pretty impressed to get even 20Mb sustained off an 8Gb disk (probably SCSI II uw? - were you fsyncing Kevin?)....but then I'd have gone with 'time cp severallargefiles somewhere' to measure performance. (yes I know that seek time is more important for database access than overall throughput, but setting up 'real world' testing is _so_ much more difficult). Colin -----Original Message----- From: marty [SMTP:[EMAIL PROTECTED]] Sent: Monday, June 24, 2002 3:46 PM To: scottish Subject: Re: [scottish] Benchmarking...how to interpet Bonnie++ output? > Of the following two sets of data, which is faster? Faster to do what? OS A is faster at handling character-at-a-time data, OS B is faster at handling block-at-a-time data. The one that is faster for you depends on whether your applications are going to do more getc/putc or more read/write. Cheers, Martin -- Martin McCarthy /</ http://www.non-prophet.org [EMAIL PROTECTED] \>\ http://www.ancient-scotland.co.uk -------------------------------------------------------------------- http://www.lug.org.uk http://www.linuxportal.co.uk http://www.linuxjob.co.uk http://www.linuxshop.co.uk -------------------------------------------------------------------- -------------------------------------------------------------------- http://www.lug.org.uk http://www.linuxportal.co.uk http://www.linuxjob.co.uk http://www.linuxshop.co.uk --------------------------------------------------------------------
