It's interesting work.
Dave,
Since I am just a young student and this is part of my thesis work on some experimental TCP enhancements implementation, I can't hide that my biggest wish when I released it was to receive feedback and advices by the linux-net developers.
So thank you a lot for the interest & precious comments.
> There is one tertiary issue I've considered wrt. this kind > of packet spacing. Consider a router that uses a routing > cache like Linux does, or something similar, and when hits occur > the hash chain member is moved to the front. Here, packet bursting > can actually benefit performance by better taking advantage of the > hash chain element having been moved to the front.
> It's just a consideration, and frankly in modern routers it probably > is pretty irrelevant. Well... most modern routers don't use the > classic FIFO/drop-tail policies which is what this patch is designed > for. :-)
You are right.
But it still looks that in some rare cases it could help.
For example, I heard that someone is actually evaluating benefits of packet spacing over some kind of wireless networks: it seems that in presence of large RTTs combined with network "blackouts", spreading packets can improve performances.
Regards
-- Daniele Lacamera - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
