Matthew Jacob wrote... > > > Well, yes, it can if the number of tagged commands you throw at a device > overloads the sequencer so it can never really get started on the first > command. If you don't disconnect, you don't throw multiple commands at > the disk. I don't remember from the original mail whether or not this was > a raw device or not.
I think he was doing sequential writes through the filesystem, using iozone and bonnie. The Seagate he was complaining about is a Medalist Pro, I think. So it's not one of their high-end disks, and could conceivably get overwhelmed by a whole lot of traffic. > On Sat, 20 Feb 1999, Matthew Dillon wrote: > > > :This did not improve anything , but I think I have found the couse. > > :In that modepage there is a DISC value which was 0 on the IBM and 1 on the > > :Seagate. I remembered a ' Enable disconnect' option in the Adaptec 2940 > > bios, > > :setting this to 'off' for both harddisks led to a huge performance > > increase on > > :the Seagate. If I also enable Ultra mode iozone write goes from 1.5 MB/s > > :to 12 MB/s ( a factor of 8 !!!). Hmm, I missed that tidbit when I first read the message. So what happens if you enable disconnection, and run at Ultra speeds? In general, you want to run with disconnection enabled. > > : Paul > > : > > :-- > > :Paul van der Zwan paulz @ trantor.xs4all.nl > > > > There's something wrong. Disconnection should not cause that sort > > of performance decrease. Disconnection is necessary if you want to > > maintain preformance with more then one scsi device on the scsi bus. > > > > -Matt > > Matthew Dillon > > <dil...@backplane.com> > > Ken -- Kenneth Merry k...@plutotech.com To Unsubscribe: send mail to majord...@freebsd.org with "unsubscribe freebsd-current" in the body of the message
