Re: [zfs-discuss] Some test results: ZFS + SAMBA + Sun Fire X4500 (Thumper)
Tim Thomas wrote: Hi this may be of interest: http://blogs.sun.com/timthomas/entry/samba_performance_on_sun_fire I appreciate that this is not a frightfully clever set of tests but I needed some throughout numbersand the easiest way to share the results is to blog. It seems that we can conclude that for this workload (streaming write over SAMBA), you saturated 2 x 1Gb/sec ethernet links, and the rest of the system (CPU, disk bandwidth) was under-utilized. --matt ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Some test results: ZFS + SAMBA + Sun Fire X4500 (Thumper)
Title: Signature Will et al I added a few extra graphs to the original posting today showing the work that an individual disk was doing http://blogs.sun.com/timthomas/entry/samba_performance_on_sun_fire and ran the RAID-Z config with fewer disks just to see what happened. http://blogs.sun.com/timthomas/entry/another_samba_test_on_sun What I find nice about Thumper/X4500's is that they behave very predictably..in my experience anyway. Rgds Tim -- Tim Thomas Storage Systems Product Group Sun Microsystems, Inc. Internal Extension: x(70)18097 Office Direct Dial: +44-161-905-8097 Mobile: +44-7802-212-209 Email: [EMAIL PROTECTED] ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Some test results: ZFS + SAMBA + Sun Fire X4500 (Thumper)
Hi Tim, Sorry for mailing you directly; I meant to reply to the list. My mistake. On 10/8/07, Tim Thomas [EMAIL PROTECTED] wrote: this was a one day project which is why I kept it simple and I don't have detailed data beyond what I collected for the graphs. The graphs are pretty informative, even so. You can see the limitations of raid-z on it pretty clearly - 200 megabytes per second requires six thousand IO/sec?! That's about 30k per I/O, which lines up with what's expected for 128k stripes on five-wide stripes... but that's still a lot of overhead. I was lucky enough to have the hardware to put this together (threw this ?) together very quickly I work in Engineering in Sun's Systems Group in an ISV focused team. We are having some discussions in the team about running a full suite of tests against SAMBA on Thumper but no firm plans yet...it is all about scheduling :-( I can certainly understand scheduling limitations, and look forward to any further results. Having local equivalents to the samba tests would also be interesting - write a local file at maximum speed, and see how fast that goes and how many I/Os that generates. for example. Non-sequential tests would be neat, too. Will ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
