Date: Tue, 19 Oct 2004 13:33:18 -0700 From: "Mitch (bitblock)" <mitch@bitblock.com> To: "Kristofer Pettijohn" <krishopper@cybernetik.net>, freebsd-isp@freebsd.org Subject: RE: Disk I/O Performance with CCD Message-ID: <OCEGLFACMGOKINMEOKANIEDEDMAA.mitch@bitblock.com> In-Reply-To: <20041019192252.GB78974@cybernetik.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Looking at the ST373453LC (a faster version of the same capacity?) couldn't find the same drive online... is your model number right? http://www.seagate.com/cda/products/discsales/enterprise/tech/0,1084,549,00. html Here are the data transfer rates... as you are running a 10K drive? you could probably assume about 0.66 of these numbers... Internal Transfer Rate (min) 632 Mbits/sec Internal Transfer Rate (max) 891 Mbits/sec Formatted Int Transfer Rate (min) 56 MBytes/sec Formatted Int Transfer Rate (max) 86 MBytes/sec So lets assume: Internal Transfer Rate (min) 417 Mbits/sec Internal Transfer Rate (max) 588 Mbits/sec Formatted Int Transfer Rate (min) 36 MBytes/sec Formatted Int Transfer Rate (max) 58 MBytes/sec This means in THEORY, you agregate transfer rate should range between 134MB - 216MB.... HOWEVER: You are reading and writing - i.e. thrashing the heads back and forth well beyond the internal drive caches' ability to absorb... You are using software raid, which probably waits on full writes and reads to each drive, which without spindle sync may have more delay related to the max latency to find a sector (related to on average about the time of 1 half of one revolution...) 160MB/s is the bus speed... with command overhead, latency, contention, etc. etc. your real throughput would be even less... not sure what else I'm missing, but I think you are setting the bar too high ;-) A hardware controller with large cache or perhaps capable of striping across channels might help increase total throughput... Just a thought. m/ > -----Original Message----- > From: owner-freebsd-isp@freebsd.org > [mailto:owner-freebsd-isp@freebsd.org]On Behalf Of Kristofer Pettijohn > Sent: Tuesday, October 19, 2004 12:23 PM > To: freebsd-isp@freebsd.org > Subject: Disk I/O Performance with CCD > > > > Im looking for some suggestions on I/O performance. > > I'm using FreeBSD 4.10-RELEASE on a Usenet transit server running > Diablo for the transit software. > > I have 4 Seagate ST373435LC SCSI drives, 70GB each, and I am using > CCD to bind them together with RAID-0 stripes. > > I can pull in anywhere from 30-40 MB sec and push out ~ 8-15 MB/sec.. > averaging about 50 MB/sec throughput.. feeds coming in are coming > in just fine, but sending stuff back out is lagging behind.. its > falling about a half hour behind every hour. > > I've used tunefs to set the average file size to 20 MB and enabled > soft-updates, as these are generally larger binary files that just > get appended to, and then seeked later on to send the article out, > I've played with setting the stripe size from anywhere between 8MB > and 64MB, and did not see much change on performance between those. > > Maybe I'm just missing something small, but on these SCSI drives > which have 160 MB/s transfer rates, I'm expecting a bit more than > I'm getting with CCD. > > Can someone give me any pointers to look at or suggestions of things > to try? > > Thanks! > > Kristofer > _______________________________________________ > freebsd-isp@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-isp > To unsubscribe, send any mail to "freebsd-isp-unsubscribe@freebsd.org" >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?OCEGLFACMGOKINMEOKANIEDEDMAA.mitch>