ietf-asrg
[Top] [All Lists]

Re: [Asrg] A New Plan for No Spam / Velocity Indicator

2003-05-01 16:20:26

Moore's law describes the growth of transistor density over time.

What has increased the volume of mail volume processing is the
dramatic fall in price of processing. A single-MIPS machine used to be
half-million dollars, now a 3GHZ machine is around one thousand
dollars, and that includes all the system components (disk, ram, etc)
in both examples.

Since mail is reasonably distributed in nature that dramatic drop in
price allows one to have 100 mail servers all working in (relative)
unison and approximately additive capacity.

However, Moore's law per se would have been satisfied if that
half-million dollar machine were still $500K, but had increased from
1MHZ to 3GHZ.

So something else has occurred, albeit arguably related to Moore's Law
if you impose a price per computron per pound rule (thus higher
transistor density should come down in price.)

I guess the two, exponentially faster/denser hardware and price drops,
have become closely associated in our heads. But it's not necessarily
case.

In fact Cray (badly) bet it would not be the case and gambled their
business on the fastest/densest hardware being extremely expensive.

Ah well. But the point is it wasn't obvious to everyone, so 20/20
hindsight and all that.

And *MY* point was that faster CPUs are only part of the picture when
viewing servers, a problem people like myself have to grapple with all
the time when some breathtakingly fast CPU'd machine won't even barely
character echo because the mail load is backed up on disk I/O usually
from a spam attack. At that moment it just doesn't feel like Moore is
smiling on me.

-- 
        -Barry Shein

Software Tool & Die    | bzs(_at_)TheWorld(_dot_)com           | 
http://www.TheWorld.com
Purveyors to the Trade | Voice: 617-739-0202        | Login: 617-739-WRLD
The World              | Public Access Internet     | Since 1989     *oo*
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg