Had to read and reread the humongous IDE vs SCSI RAID
thread 3-4 times over to fully appreciate the issues...

"[UDMA 133] is a mirage and an irrelevancy.  That's the 
theoretical maximum transfer rate of the bus, which can 
never be approached even in ideal cases, because data flow 
is limited to the physical read speed of hard drives,"

...is probably untrue because the presence of large caches 
on today's IDE hard drives (I believe they're up to 4 to 8MB 
on high end drives nowadays) does make a practical difference
in the throughput of hard disks. As to whether Linux's 'efficient 
cache buffer management' will necessarily 'mask the effect' of 
onboard hard disk caches, I think that's an open question.  The 
best way to settle this would be to disable the onboard hard disk 
caches and do a benchmark. Now taking bets... :-)

Now about SCSI disconnect... how does that work and how does it
allow "bandwidth aggregation" among multiple SCSI devices on the
same chain when at any given time, on a shared bus model, only one 
device can be transmitting data at a time?  At best, it sounds like
just a finer grained relinquishing of the bus. Thus, with a 2
drive setup (the maximum under ATA), it might not make a big
difference in performance, certainly not enough to justify the
huge price difference of a SCSI drive.

_
Philippine Linux Users Group. Web site and archives at http://plug.linux.org.ph
To leave: send "unsubscribe" in the body to [EMAIL PROTECTED]

To subscribe to the Linux Newbies' List: send "subscribe" in the body to 
[EMAIL PROTECTED]

Reply via email to