Seagate Barracuda SCSI-II drives
: In <3jl5qc$...@nyx.cs.du.edu> jhank...@nyx.cs.du.edu (jonathan hankins) writes:
: >Has anyone had any successes/failures with the new Seagate barracuda
: >high-rpm drives? I didn't really think the higher rpms would be a
: >variable, but I wanted to make sure before we buy one ;-)
: I've been using a segate 32550N (barracuda 2lp) for a month or so now,
: and it works great. The only real dissapointment is that even though
: the drive is twice as fast and has twice the number of sectors per track
: as my 5 year old maxtor, I only get marginally better performance under
: Linux (I'm getting about 1.3 MB/s read and 900kb/s write). According
: to my calculations, the drive should be capable of > 5mb/s (109 sectors/track,
: 7200rpm = 109 * 512 * 120rps = 6.4mb/s). This is all on a P5-90 with a
: 1542CF and linux 1.2.0 on an ext2 fs. Under dos, my ancient version of
: coretest claims 2mb/s read.
I suggest that there may be a bottleneck in the SCSI card or the
SCSI setup. Perhaps moving to a PCI or VLB SCSI card (like a Buslogic or
something) would lower the CPU bottleneck considerably.
A Barracuda should run better than that, and I can only suggest
something at the interface level...
: If anyone has any ideas as to why performance is so poor, I'd love to hear
: (I realize that going to a PCI controller would help - any ideas on what
: a generic NCR53c810 would do for me?)
: One word of caution on the barracuda (and probably other 7200 rpm drives) is
: that they run VERY hot. I ended up putting the drive in a bay with a fan
: blowing air over it to keep it cool (and I've got one of the new "cooler"
: running drives).
Well, they are hot drives... :)
- Chad Page
: -Roger
: --
: ---------------- | Death to XMM!
: Internet: sunsh...@convex.com | Why does Tempest run so poorly
: UUCP: ...!uunet!convex!sunshine | under Windows on a 386/25 when it
: ATTNET: (214)497-4779 | works so well on a 6502?