High-fidelity simulation of load balancing methods in Tor

We evaluate the effectiveness of two novel performance improving algorithms for Tor using the Shadow simulation framework. The first is Tightrope, a path selection algorithm for improving the load balancing of Tor circuits across the Tor network. The second is using the maximum likelihood estimation (MLE) scheme to accurately estimate the true capacity of relay capacities with minimal prior knowledge. For both algorithms, we presented the technical implementation details and the experiment setup. In Tightrope's simulation results, we found that the new path selection method is effective in improving the overall throughput of the network. In MLE's simulation results, we verified its estimation accuracy under different settings.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Electrical & Computer Engr

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2020

Author dc:creator

• Cheng, Chin-Yu

Contributors dc:contributor

• Borisov, Nikita

• Distributed systems, Tor, networking, load balancing

Statement dc:rights

• Copyright 2020 Chin-Yu Cheng

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/108201

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/108201