Beowulf: A theorical approach
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Greg Lindahl glindahl at hpti.comThu Jun 22 11:10:32 PDT 2000
- Previous message: Beowulf: A theorical approach
- Next message: Beowulf: A theorical approach
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
> Just as a matter of curiosity -- Once upon a time some two or three > years ago I suggested on the list that a development company consider > building a network communications device that plugged into the second > CPU slot of a dual CPU board. It's hard to build something that plugs into an Intel-designed CPU bus. Scali tried it and it didn't work so hot. You could do it for the Alpha/AMD cpu bus, but it's a high speed design, and the estimates I got (I was seriously considering this) was 12 months and $2 million. In 12 months, you'll be able to get either PCI-X machines or Infiniband machines. PCI-X has split transactions, and Chuck Seitz, who knows more about networking than all of us combined, says that PCI-X will get rid of most of the PCI latency. So, sit back, pop open a cold one, and wait. -- greg
- Previous message: Beowulf: A theorical approach
- Next message: Beowulf: A theorical approach
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
