| To: | Stephen Hemminger <shemminger@xxxxxxxx> |
|---|---|
| Subject: | Re: [RFC] TCP Vegas for 2.6 |
| From: | jamal <hadi@xxxxxxxxxx> |
| Date: | 09 Mar 2004 13:58:06 -0500 |
| Cc: | John Heffner <jheffner@xxxxxxx>, Andi Kleen <ak@xxxxxxx>, linux-net <linux-net@xxxxxxxxxxxxxxx>, netdev@xxxxxxxxxxx |
| In-reply-to: | <20040309102206.33c1720c@dell_ss3.pdx.osdl.net> |
| Organization: | jamalopolis |
| References: | <20040309180331.GC11604@wotan.suse.de> <Pine.NEB.4.33.0403091307090.5230-100000@dexter.psc.edu> <20040309102206.33c1720c@dell_ss3.pdx.osdl.net> |
| Reply-to: | hadi@xxxxxxxxxx |
| Sender: | netdev-bounce@xxxxxxxxxxx |
On Tue, 2004-03-09 at 13:22, Stephen Hemminger wrote: > Every case I tested has vegas faster than the default reno. It is especially > noticeable over the DSL. ] As i am sure you will notice it even more on ppp/dial up. I am suprised on the DSL results though. Theres tons of literature has already been written on this stuff. It is know that Reno will work better over Vegas in some cases (as pointed above). Its useful to have vegas as a feature but lets not jump the gun. Congestion control algorithms have been known to start religious wars. > The current implementation is not ready to be turned > on by default though. i like Andis idea of being able to select per route. cheers, jamal |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [PATCH] Trivial kmalloc failure checks, David S. Miller |
|---|---|
| Next by Date: | Re: ip addr flush hangs, David S. Miller |
| Previous by Thread: | Re: [RFC] TCP Vegas for 2.6, Nivedita Singhvi |
| Next by Thread: | Re: [RFC] TCP Vegas for 2.6, Nivedita Singhvi |
| Indexes: | [Date] [Thread] [Top] [All Lists] |