Re: Possible TCP Problem with RH6.2 talking to Solaris2.6/2.7

From: James Sutherland (jas88@cam.ac.uk)
Date: Sun May 07 2000 - 14:55:41 EST


On Sun, 7 May 2000, Richard Gooch wrote:
> kuznet@ms2.inr.ac.ru writes:
> > Hello!
> >
> > > That's a *very* high load percentage. We put over 20MB/sec through old P100
> > > systems with two fast ethernets channel bonded using Tulip 21140 cards on
> > > 2.0.30, and it put about that load (slightly less) on the system.
> >
> > Well, all the things have tendency to degrade with time.
> >
> > But what memory bandwidth did this old P100 have? Even if to reduce transfer
> > to one plain bcopy, you had >60MB/sec. I daresay it was pretty mighty P100. 8)
> > Seems, hardware also degrades.
>
> ??? Hardware slows down over time?!? The CPU starts taking 5 cycles to
> do something that it used to do in 4?!?
> Surely you must mean something else.

He was disbelieving the figures quoted, which would (he thought) have
implied a machine faster than current ones.

In fact, 60Mb/sec on a 64 bit bus implies 7.5MHz, or 133ns memory. For a
P100, that would have been a little slow...

James.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Sun May 07 2000 - 21:00:21 EST