| To: | Baruch Even <baruch@xxxxxxxxx> |
|---|---|
| Subject: | Re: [PATCH 2/5] TCP BIC 1.1 support |
| From: | Stephen Hemminger <shemminger@xxxxxxxx> |
| Date: | Thu, 7 Apr 2005 12:02:38 -0700 |
| Cc: | Andrew Morton <akpm@xxxxxxxxx>, "David S. Miller" <davem@xxxxxxxxxxxxx>, netdev@xxxxxxxxxxx, Injong Rhee <rhee@xxxxxxxxxxxx> |
| In-reply-to: | <423B9415.2010906@ev-en.org> |
| Organization: | Open Source Development Lab |
| References: | <20050318162211.366ca490@dxpl.pdx.osdl.net> <423B9415.2010906@ev-en.org> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
On Sat, 19 Mar 2005 02:53:09 +0000 Baruch Even <baruch@xxxxxxxxx> wrote: > Stephen Hemminger wrote: > > This patch adds TCP BIC back in as a pluggable TCP congestion mechanism. > > This version is closer to the TCP BIC 1.1 released for Web100. > > The changes from 2.6.11 are: > > * congestion window undo fix > > * delayed ack compensation The delayed ack calculation code in this patch sequence doesn't work quite right. It over estimates the number of delayed acks and this causes BIC to clamp down too hard. I'll look into it later, but just wanted to let people know that it this version of BIC performs worse under delay. The problem is in the calculation of ack_ratio not in the BIC portion. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [PATCH] Too aggressive cwnd backoff, John Heffner |
|---|---|
| Next by Date: | Re: [PATCH] netem: account for packets in delayed queue in qlen, Stephen Hemminger |
| Previous by Thread: | [PATCH] Too aggressive cwnd backoff, Baruch Even |
| Next by Thread: | Re: [PATCH 2/5] TCP BIC 1.1 support, Stephen Hemminger |
| Indexes: | [Date] [Thread] [Top] [All Lists] |