Re: FTP benchmark proposal

Oliver Xymoron (oxymoron@waste.org)
Sun, 27 Jun 1999 17:37:55 -0500 (CDT)


On 27 Jun 1999, Andi Kleen wrote:

> lm@bitmover.com (Larry McVoy) writes:
> >
> > with each of the "c" machines masquerading as a pile of clients. The
> > reason for that is that we want to generate the real pcb lookup load
> > on the server so it has to think that it is talking to 6000 clients.
>
> The hard thing will be to simulate slow/timing out connections that occur over
> the internet. Just low-latency local networking with no lost packets is easy.
>
> So perhaps something like a NISTNET box is needed inbetween?

How about hacks to the traffic shaper on the client machines? Choose some
distribution function for bandwidth and latency for clients and randomly
generate pipes that accommodate said parameters. Right now, of course,
shaper only does bandwidth, since no one in RL wants to add latency, but
it should be doable without too much pain. We can also make shaper do a
more realistic simulation of congestion to generate stalls, etc.

--
 "Love the dolphins," she advised him. "Write by W.A.S.T.E.." 

- 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/