On 25 Nov 1996, D. J. Bernstein wrote:

> ``qmail beats remote SMTP servers to death'': Nonsense. The concept of
> this type of overload is ludicrous. Even the _maximum theoretically
> possible_ burst---which will never actually happen outside controlled
> tests---won't hurt a remote 486 running qmail.

Unfortunately, this neglects the fact that there are many SMTP
servers out on the Internet that are not running qmail. A number
of these are running terribly bad server software on Windows NT
machines, and you can bet that this proportion is only going to
increase.

I do know of sites where if every mailing list coming in were
running on a qmail server, the load server would have a distinct
possiblity of being brought to its knees.

The possibility of this, and even the small amount of added bandwith
that qmail adds, strike me as an excessive cost to pay for having
a piece of e-mail delivered anywhere from a tenth of a second to
a few seconds earlier. I just can't see the benefit of this marginally
earlier delivery at all, except to performance freaks for whom
benchmarks are more important than reality.

cjs

Curt Sampson    cjs@portal.ca		Info at http://www.portal.ca/
Internet Portal Services, Inc.	
Vancouver, BC   (604) 257-9400		De gustibus, aut bene aut nihil.