[or-cvs] r18777: {projects} add three more speed-up ideas (projects/performance)

[arma at seul.org]

Author: arma
Date: 2009-03-05 05:57:22 -0500 (Thu, 05 Mar 2009)
New Revision: 18777

Modified:
   projects/performance/performance.tex
Log:
add three more speed-up ideas


Modified: projects/performance/performance.tex
===================================================================
--- projects/performance/performance.tex	2009-03-05 10:54:58 UTC (rev 18776)
+++ projects/performance/performance.tex	2009-03-05 10:57:22 UTC (rev 18777)
@@ -64,17 +64,21 @@
 This document describes our current understanding of why Tor is slow,
 and lays out our options for fixing it.
 
-The Tor network is slow right now for six main reasons, with the most
-severe listed first. For each reason, we explain our current intuition
-for how to fix it, how effective we think it would be, how much effort
-and risk is involved, and the recommended next steps.
+The Tor network is slow right now for six main reasons. We discuss each
+reason in its own section, starting with the most severe. For each reason,
+we explain our current intuition for how to fix it, how effective we
+think each fix would be, how much effort and risk is involved, and the
+recommended next steps.
 
 \pagebreak
 \tableofcontents
 \pagebreak
 
-\section{Congestion control not good}
+\section{Our congestion control does not work well}
 
+Tor's most critical performance problem is in how it combines high-volume
+streams with low-volume streams. ...
+
 \subsection{TCP backoff slows down all streams since we multiplex}
 
 End-to-end congestion avoidance
@@ -172,6 +176,7 @@
 Therefore, a different optimum may exist for networks with different
 characteristics.
 
+\subsection{Priority for circuit control cells, e.g. circuit creation}
 
 
 \section{Some users add way too much load}
@@ -426,9 +431,11 @@
 \section{Better handling of high/variable latency and failures}
 
 \subsection{The switch to Polipo: prefetching, pipelining, etc}
-\subsection{bad timeouts for giving up on circuits and trying a new one}
+\subsection{better timeouts for giving up on circuits and trying a new one}
+\subsection{when a circuit has too many streams on it, move to a new one}
 \subsection{If extending a circuit fails, try extending a few other
 places before abandoning the circuit.}
+\subsection{Bundle the first data cell with the begin cell}
 
 
 \section{Network overhead too high for modem users}