Scaling asynchronous multi-party computation: A systems perspective

Modern multi-party computation applications no longer have a one-time execution pattern and instead are required to be run continuously like a service. They are deployed over the Internet which is inherently asynchronous and demand an infrastructure which is end-to-end robust, fault-tolerant and scalable. Unfortunately, existing frameworks fail to satisfy all of these requirements. Hence, many MPC applications are not yet practical due to the lack of an MPC framework that meets these needs. This work presents a scalable protocol for generating preprocessed elements required for the execution of asynchronous MPC applications with optimal Byzantine fault-tolerance (robust when one-third of the nodes are corrupt) in the asynchronous setting. We implement this preprocessing protocol in HoneyBadgerMPC – a scalable, robust and fault-tolerant framework designed to develop, test and benchmark MPC applications efficiently.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2019

Author dc:creator

• Kulshreshtha, Samarth

Contributors dc:contributor

• Miller, Andrew E

• Systems
• Multi-party Computation
• Cryptography

Statement dc:rights

• Copyright 2019 Samarth Kulshreshtha

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/104886

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/104886