* Posts by Matt Reilly

2 posts • joined 14 Mar 2009

Ethernet — a networking protocol name for the ages

Matt Reilly

Metcalfe and Contributions

OK. First, a disclaimer: I know Bob Metcalfe. He was the Chairman of the board of a company I founded. He's still a member of the board.

Suggesting that his contribution was inferior to other approaches, or "didn't work" or whatever is patently absurd, but it also misses a major point: Ideas are cheap -- lots of people have interesting ideas, from the guy who wanders up and down the street talking about free energy from his inter-stellar visitor friends to Nobel prize winners. Ideas are cheap. Turning them into reality takes more than most people have: ambition, drive, resilience, and often charisma.

Ethernet at the start was not a slam dunk. (I have a copy of the note from Robert Bachrach at PARC completely trashing the idea of Ethernet and CSMA.) Bringing three different vendors together was not a slam dunk. Building a new company is never a slam dunk.

Folks who build stuff take risks. Some of the risks may seem DOA in hindsight. Success is never a slam dunk.

However, posting anonymously is always a slam dunk: nothing risked. (Unless your mommy happens to open the door.)


Supercomputer niche chucks rocks at Nehalem

Matt Reilly
Thumb Up

Mem and Comm BW

Michael H.F. Wilkinson states it clearly and precisely: what matters is bandwidth -- both between processes and between the arithmetic unit and memory. It isn't about raw FLOPS, GHz, or IntelInside.

(And in fact, bandwidth and power is the whole point of the SiCortex architecture, though in its setting applications routinely scale to hundreds and even thousands of processors.)

In all the time I spent talking to customers and prospects, the choice of the instruction set (MIPS vs. x86) was never much of an issue.