home *** CD-ROM | disk | FTP | other *** search
- Path: news.dcs.warwick.ac.uk!str-ccsun!not-for-mail
- From: nbc@vulture.dmem.strath.ac.uk (Neil Brendan Clark)
- Newsgroups: comp.sys.amiga.advocacy,comp.sys.amiga.misc
- Subject: Re: New Press Release!
- Date: 19 Mar 1996 23:59:11 -0000
- Organization: University of Strathclyde
- Message-ID: <4inhof$g1v@vulture.dmem.strath.ac.uk>
- References: <4hon0p$rfm@gidora.kralizec.net.au> <4ibtk4$49a@hubcap.clemson.edu> <4icgak$rt6@serpens.rhein.de> <4icl5c$hek@hubcap.clemson.edu>
- NNTP-Posting-Host: vulture.dmem.strath.ac.uk
-
- Charles E Taylor IV <charlet@hubcap.clemson.edu> wrote:
- >
- >It was misspelled, but are you trying to tell me that a faster processor
- >won't speed up operations like archiving/dearchiving?
-
- It won't after a certain point in the CPU/disk speed ratio. In other words,
- when one reaches the point where they can pump out data from archives at
- a rate equal to that which the disk can accept it, then any extra CPU
- speed gives you no benefit. Of course, this is ignoring the effects of
- disk caches and the like. Refer to Ahmdal's (sp?) Law in Hennessey and
- Patterson, "Computer Architecture: A Quantitative Approach" (approximately).
-
- Apparently the fact that CPU speeds have risen far more quickly than memory
- and I/O device speeds has become a real problem of late in the computer
- industry.
-
- --
- "Ye'd better kiss King Willie's loof, than come by Killiecrankie-o", The Corries
-
- Neil Clark, Transparent Telepresence Group
- <http://telepresence.dmem.strath.ac.uk/>
-