Weaving a Faster Tor: A Multi-Threaded Relay Architecture for Improved Throughput

Abstract

The Tor anonymity network has millions of daily users and thousands of volunteer-run relays. Increasing the number of Tor users will enhance the privacy of not just new users, but also existing users by increasing their anonymity sets. However, growing the network further has several research and deployment challenges. One such challenge is supporting the increase in bandwidth required by additional users joining the network. While adding more Tor relays to the network would increase the total available bandwidth, it requires network architecture changes to reduce the impact of Tor’s growing directory documents. In order to increase the total available network bandwidth without needing to grow Tor’s directory documents, this work provides a multi-threaded relay architecture designed to improve the throughput of individual multi-core relays with available network capacity. We built an implementation of a subset of this new design on top of the standard Tor code base to demonstrate the potential throughput improvements of this architecture on both high- and low-performance hardware.