On Wed, Jul 11, 2001 at 12:07:50PM +0100, Colm Buckley mentioned:
>> I've used IDE and SCSI systems extensively over the past few years,
> and they both have merits. Chief among IDE's merits is cost of
> purchase; the mass-market economics of the controllers and drives have
> really driven costs down.
>> However, cost of purchase isn't the only cost associated with a
> system - my experience of IDE is "it's only cheaper if your time is
> free". Modern SCSI systems "just work", and you don't have to waste
> time messing around and worrying about whether DMA is on, or which
> transfer mode your controller is in.
>> Also, you can have up to sixteen devices per SCSI bus, including the
> controller. Big win if you want to do RAID and whatnot.
Not as much as you'd think. 80MB/sec is the max for decent SCSI
(160MB/sec exists but it rare), and 80MB/sec can be saturated by a 3+1
RAID 5 array of anyway decent disks.
> Unlike IDE, the slowest device on the bus doesn't hamper the whole
> bus.
Right and wrong. Connect a narrow/slow scsi device to a fast/wide bus
and it will kill performance. 3MB/sec off such a device will knock
50MB/sec off the overall bus throughput. What won't matter on SCSI is
access latency though, which is normally the more important. Though IDE
supports disconnects in the latest versions AFAIK.
> Unlike IDE, more than one device can be active at a time. Transfers
> between devices need not involve the controller at all.
In theory, but never used in practice.
> Unlike IDE, devices can be dynamically added and removed.
Some IDE controllers support this.
> Unlike IDE, commands can be queued up at the controller and the CPU
> need not become involved until they're all finished.
The benefit of this is fairly limited in practice, and its also now
available on IDE.
> In a situation where you have both a large data flow *and* intense CPU
> activity (eg : a big compile), SCSI really wins - the constant
> mode-switching which the CPU has to do to cope with IDE really hits
> performance. Example, my machine ogma (fairly modest PIII/500) takes
> about fifteen minutes to compile the kernel on its EIDE drive (ATA66,
> DMA enabled), but only eleven on its SCSI drive.
10K SCSI disk Vs less on IDE, most likely.
> Basically, people who go on and on about how SCSI is better are
> sometimes just snobs repeating what they've been told, but at other
> times (like this), they're speaking from real experience. I wouldn't
> use IDE on a machine where disk I/O or ongoing maintenance was going
> to be anything other than an trifle.
Using an identical spec on IDE, the difference between a single device
on IDE and SCSI is minimal. SCSI has its advantages on multiple devices,
but since 4 ide channels are common enough these days, SCSI is hard to
justify on a cost basis. And if i really need a pile of storage space, i
would still consider raidzone arrays, which are IDE based, albeit 16
channel controllers.
Tony.
--
Tony Bolger Tony.Bolger at palamon.ie
Palamon Technologies Ltd. +353 86 856 2525
Maintained by the ILUG website team. The aim of Linux.ie is to
support and help commercial and private users of Linux in Ireland. You can
display ILUG news in your own webpages, read backend
information to find out how. Networking services kindly provided by HEAnet, server kindly donated by
Dell. Linux is a trademark of Linus Torvalds,
used with permission. No penguins were harmed in the production or maintenance
of this highly praised website. Looking for the
Indian Linux Users' Group? Try here. If you've read all this and aren't a lawyer: you should be!
