toad wrote:
> On Sun, Dec 10, 2006 at 09:54:59PM +0100, Jano wrote: >> Jano wrote: >> >> >> Looking forward to it! Backoff alone seems to get good throughput at >> >> high loads, but with a poor success rate (see the attached graphs, >> >> averaged over three runs) - I'll be interested to see whether LIFO's >> >> high success rate can be combined with backoff's high throughput. >> > >> > I have a new run going from 5 to 25 that I'll try to post this night. >> > We'll see what it shows... >> >> Here are these results. I'd like to hear your comments, since the clear >> LIFO advantage is curious (mind you, in some cases it more than doubles >> the other techniques) and maybe it's a simulator artifact. Do we know >> what's the typical route length in these simulations? I expected the >> multi-hop thing to ruin the intuitively good performance of the >> single-hop case. >> >> Here's a svn diff, in case you see something wrong with my changes: > > My guess would be we're not counting the requests that never complete? This is true, and is funny because I was just thinking of it. This obviously skews the ratios optimistically. The total throughput however should be accurate, and this is more relevant I'd say. However this is not consequence of my changes; the other simulations are affected as well...
