home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.sgi
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!sgiblab!sgigate!sgi!classic.asd.sgi.com!jeremy
- From: jeremy@classic.asd.sgi.com (Jeremy Higdon)
- Subject: Re: Exebyte 8500 config on a Crimson
- Message-ID: <s8lvg64@sgi.sgi.com>
- Sender: jeremy@classic.asd.sgi.com
- Organization: Silicon Graphics, Inc. Mountain View, CA
- References: <1992Nov10.172707.29397@seismo.do.usbr.gov>
- Date: Thu, 12 Nov 1992 05:19:27 GMT
- Lines: 36
-
- In article <1992Nov10.172707.29397@seismo.do.usbr.gov>, geomagic@seismo.do.usbr.gov (Dan O'Connell) writes:
- > We installed an Exebyte 8500 on a Crimson running 4.0.5C. It has
- > a Peripheral Visions compression board with a front door display
- > panel. We have make all the various devices for it; tps1d4, tps1d4.8200,
- > tps1d4.8500, tps1d4nr, tps1d4nr.8200, tps1d4nr.8500, etc.
- > When we write to using tar, we never get a transfer rate exceeding
- > 120 Kbytes/sec. We have tried both 8500 and 8200 modes, with and without
- > compression. We get about 180 Kbytes/sec writting to our 4mm DAT drive.
- >
- > We expected to get 300-400 Kbytes/sec transfer rates with the Exebyte
- > in 8500 mode. Is there something obvious we should look for that would
- > explain the poor Exebyte performance?
- >
- > Here is the info returned by mt
- >
- > mt -t /dev/rmt/tps1d4 status
- > Controller: SCSI
- > Device: EXABYTE: EXB-8500-85Qanx00446
- > Status: 0x200
- > Drive type: 8mm cartridge
- > Media : Not READY
- >
- > Thanks for any info.
- > --
- > Dan O'Connell
- > geomagic@seismo.do.usbr.gov
- > Seismotectonics Group, U.S. Bureau of Reclamation
-
- You should test performance of the Exabyte by using 'dd' to transfer
- a large amount of data with a large block size to the no-swap device.
- With 'tar', the Exabyte may be doing a lot of start/stop, which
- is very inefficient on the Exabyte. So to minimize other differences,
- try 'dd'. It should have no problem streaming at nearly 500KB/s.
-
- jeremy
-
-