🚀 go-pugleaf

In article <181@tijc02.UUCP> seb022@tijc02.UUCP (Scott Bemis) writes:
> I currently am using two CDC 9772 disk drives [...] Carefully controlled
> benchmarks showed the these disk drives to be 43 percent faster than the
> RA81 disk drives.

	The 9772's may indeed be a very good drive (I don't know, I've
never seen one), but to claim that it is fast just because it beats the
pants off an RA-81 is absurd.  The RA-81's are slow by any modern standard.
According to the manufacturer's specs, the RA-81 has a 28 ms average seek
time (actually, they quote it as "average positioning rate", whatever that
means).  The Fuji 2351 Eagle, for example, quotes 18 ms.  I only pick the
2351 because it is of about the same vintage as the RA-81, both being
introduced about 3-4 years ago or so.  The difference is that the Eagle is
still considered a reasonable drive to buy (although clearly the last few
years have given it some competition) while the RA-81 was old when it was
introduced.  You can also buy Eagles (and, I suspect, most SMD disks) for
about half the price per Mbyte of an RA-81.

	To be fair, the RA-81 was a radical departure in disk subsystem
design.  A lot of the smarts that used to be in the kernel-resident driver
are now in the controller.  I don't know how much the Unix drivers take
advantage of the ability of the controllers (such as the most common
UDA-50) to optimize disk operations.  I suspect very little.  With smarter
and smarter controllers (such as the HSC-[57]0), and drivers designed to
take full advantage of the hardware the RA-81 (and RA-82) may indeed come
into their own.  I tend to doubt it however.
-- 
Roy Smith, {allegra,cmcl2,philabs}!phri!roy
System Administrator, Public Health Research Institute
455 First Avenue, New York, NY 10016