[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

[or-cvs] r18853: {projects} section 4.3 (projects/performance)



Author: arma
Date: 2009-03-10 07:35:45 -0400 (Tue, 10 Mar 2009)
New Revision: 18853

Modified:
   projects/performance/performance.tex
Log:
section 4.3


Modified: projects/performance/performance.tex
===================================================================
--- projects/performance/performance.tex	2009-03-10 11:03:48 UTC (rev 18852)
+++ projects/performance/performance.tex	2009-03-10 11:35:45 UTC (rev 18853)
@@ -959,14 +959,18 @@
 
 Micah Sherr is finishing his PhD thesis at Penn under Matt Blaze,
 exploring
-exactly this issue. He suggests to use a virtual coordinate system --
+exactly this issue. In the past we've avoided any sort of path selection
+algorithm that requires pairwise measurements of the network, because
+communicating $N^2$ measurements to clients would take too much bandwidth.
+Micah solves this problem by using a \emph{virtual coordinate system} --
 a three or four dimension space such that distance between relays in
 the virtual coordinate space corresponds to the network latency (or other
 metric) between them.
 
 His experiments show that we could see a significant speedup in the Tor
 network if users choose their paths based on this new relay selection
-algorithm.
+algorithm. More research remains, of course, but the initial results
+are very promising.
 
 %A second option would be to actually measure hop latency, and publish
 %the database.
@@ -984,7 +988,8 @@
 %is an approximation of the latency of a hop between them.
 %Delta compression could be used if the map changes slowly.
 
-Reducing the number of potential paths would also have anonymity
+On the other hand, reducing the number of potential paths would also
+have anonymity
 consequences, and these would need to be carefully considered.
 For example, an attacker who wishes to monitor traffic could create
 several relays, on distinct /16 subnets, but with low latency between them.
@@ -996,11 +1001,19 @@
 
 {\bf Impact}: Medium-high.
 
-{\bf Effort}: Medium-high, since 
+{\bf Effort}: Medium-high, since we first need to sort out how effective
+the algorithm is, and then we need to figure out a migration plan.
 
-{\bf Risk}:
+{\bf Risk}: Medium, since a new selection algorithm probably carries
+with it a new set of anonymity-breaking papers that will only come out
+a few years after we deploy.
 
-{\bf Plan}:
+{\bf Plan}: Micah is going to write a design proposal for getting
+relays to compute and maintain their virtual coordinates based on
+latency. Once we deploy that, we'll have some actual data points, and
+we'll be in a better position to simulate whether the idea will help
+in reality. Counting deployment time, that means we probably won't have
+clients using this scheme until 2010.
 
 \subsection{Considering exit policy in relay selection}