Messages in this thread | | | Date | Mon, 24 Jan 2000 08:46:57 -0800 (PST) | From | David Lang <> | Subject | Re: Interesting analysis of linux kernel threading by IBM |
| |
On Sun, 23 Jan 2000, Sean Hunter wrote:
> On Sun, Jan 23, 2000 at 03:50:32PM +0100, Davide Libenzi wrote: > > We can write this : > > > > TS_old = Ko + O( N ) > > TS_new = Kn + O( log( N ) ) > > > > Where N is the RQ size. > > > > Now the curve of TS_new( N ) goes down ( intersect ) the curve TS_old( N ) in a > > point that in the worse case I've measured is N = 8 ( I prefer always to report > > worse cases to avoid to be shooted ), but I've measured even 4 with a medium > > that I can think to be near to six. > > In other words, almost all real-world cases lose. Nice... > > Sean
If you are on a 8 way SMP box a load of 4 is very reasonable. on anything less then a 4-way SMP box it is almost certinly a loss.
David Lang
- 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/
| |