mailing list archives
Re: Jumbo Frames (was Re: MAE-EAST Moving? from Tysons corner toreston VA. )
From: Bennett Todd <bet () rahul net>
Date: Mon, 19 Jun 2000 13:02:17 -0400
-----BEGIN PGP SIGNED MESSAGE-----
2000-06-19-12:19:20 Roeland Meyer (E-mail):
Actually, my testing shows a falure to utilize even 100baseTX
I'm unsurprised. For most purposes I continue to spec simple
100baseT for server<-->switch connects, 10baseT for normal clients,
and quad-100baseT etherchannel for the occasional really badass
server (e.g. a NetApp).
I don't have a 10gig-E system, but I wonder about going there when
I can't even get gig-E to work efficiently.
Well, that's you --- and me too. But that's sure not most customers.
Even some folks whose expertise I generally respect have completely
bought into gig-E, and try to apply it for host connects, without
attempting to measure whether the host is capable of saturating even
100BaseT with their traffic.
If vendors want to sell 10gig-E they should be concerned about
exactly this point.
Not true, no more than the belief that if vendors wanna be able
to charge more for GHz CPUs, they must have sufficiently balanced
systems so those CPUs really get work done faster than the
previous generation. Turns out they don't; the vast majority, that
manufacturers care about, that determine the success or failure
of a product or marketing strategy, those masses don't care about
measured performance, they care about bigger numbers and proud
boasts. RAMBUS, anyone?
Joe SOHO isn't going to buy it anyway. Joe Enterprise isn't going
to spend the extra money unless he can see some real benefit, and
Joe dot-com ain't going to do it unless it is measurably faster
than gig-E (which it won't be with MTU=1500).
I disagree once again. Joe Enterprise buys whatever the last
salescritter to buy him lunch-with-drinks tells him to buy. No other
explanation for all those "SANS" they're inflicting upon themselves,
poor boobs. Maybe Joe dot-com will be forced to get a little smarter
if his Wall St. bank stays cagey, but I wouldn't bet on that
outcome; I figure it won't be long before the dotcoms once again
have far, far more money than clue to use it, and so buy whatever
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.0 (GNU/Linux)
Comment: For info see http://www.gnupg.org
-----END PGP SIGNATURE-----