Author: Vincent Diepeveen
Date: 21:13:45 07/08/03
Go up one level in this thread
On July 08, 2003 at 23:03:01, Jay Urbanski wrote: >On July 07, 2003 at 23:35:45, Robert Hyatt wrote: > >>>I have PVM running on our giganet switch, which is faster than myrinet. But, >>as I said, such clusters are _rare_. TCP/IP is the common cluster connection, >>for obvious reasons. And that's where the interest in clusters lies, not >>in how exotic a combination you can put together, but in what kind of >>performance you can extract from a common combination. > >Giganet is not faster than Myrinet - it's 1.25Gb/s compared to Myrinet's 2Gb/s >and it has higher latency. Giganet is also no longer being sold - it's a dead >technlogy. But such clusters aren't *that* rare - I count 57 Linux clusters >with fast (better than GigE) on the TOP500 list. > >Heck - if we had a decent MPI chess program available I bet any number of those >"exotic" clusters would sign up for an exhibition match with one of the >super-GM's. One thing they all have in common is that they *love* publicity. You are wrong. Converting DIEP to MPI is very trivial. Depending upon their latency it can run then up to 500 processors. I have chatted to many cluster guys in my life. I always ask the same to them when they say: "you should run at my cluster". I say then: "give me a login please and it'll run soon very fine at your thing, because it is designed for slow latencies nowadays". So far none of them gave me a login there. If your dudes pays for a match against a top GM, I'll have a working version for you next week over MPI. Just say when you need it. I'll have it. But first show me realism behind the statement. So far a guy or 20 told me already about their clusters and you're actually the only one that keeps posting (take me right i do appreciate those posts). But this one doesn't make any sense.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.