Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ukma!rutgers!apple!amdahl!littauer
From: littauer@uts.amdahl.com (Tom Littauer)
Newsgroups: comp.arch
Subject: Re: Drum memory
Message-ID: <0fsqPccLkF1010aRTpU@amdahl.uts.amdahl.com>
Date: 24 Mar 89 14:29:52 GMT
References: <21976@lll-winken.LLNL.GOV> <28411@ucbvax.BERKELEY.EDU> <1487@wpi.wpi.edu> <95741@sun.Eng.Sun.COM>
Reply-To: littauer@amdahl.uts.amdahl.com (Tom Littauer)
Organization: Amdahl Corporation, Sunnyvale CA
Lines: 23

In article <95741@sun.Eng.Sun.COM> jamesa@arabian.Sun.COM (James D. Allen) writes:

A good summary of the use of drums on 370-architecture machines.

A minor nit: these "drums" were implemented as head-per-track disks for
ease of manufacture (cost) and capacity.

>	Data-intensive applications often read their data in large chunks.
>	A 64k chunk takes 22 milliseconds to read even at 3 MBS datarate.
>	For this reason, datarate becomes as important as seek latency
>	for these high-powered users and this tends to reduce the advantage
>	of fixed-head disks.

Correct. The need to improve on this is one of the things that drove us
to introduce 4.5MB channels and a 4.5MB data rate on our solid-state "disk".
-- 
UUCP:  littauer@amdahl.amdahl.com
  or:  {sun,decwrl,hplabs,pyramid,ames,uunet}!amdahl!littauer
DDD:   (408) 737-5056
USPS:  Amdahl Corp.  M/S 337,  1250 E. Arques Av,  Sunnyvale, CA 94086

I'll tell you when I'm giving you the party line. The rest of the time
it's my very own ravings (accept no substitutes).