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 12:29:55 PDT 2000
- Previous message: Beowulf: A theorical approach
- Next message: Beowulf: A theorical approach
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
> > It's hard to build something that plugs into an Intel-designed CPU bus. > > Scali tried it and it didn't work so hot. > > At least on Intel hardware, SMP systems have cache coherency in > the hardware, Scali's design (and the one I was considering) do not do cache coherency. They are for message passing systems. As you may gather, I'm a message passing bigot. Er, evangelist. Doing cache coherency would be like the SGI O2000, and that would be far more expensive than what I was talking about. There are approaches that attack that problem -- see the Isotach research group's work. They are at UVa, which is why I am aware of them. But not my cup of tea. -- 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
