More about HKUST
Scalable Algorithms for Gas Analysis and Optimization in Smart Contracts
The Hong Kong University of Science and Technology
Department of Computer Science and Engineering
PhD Thesis Defence
Title: "Scalable Algorithms for Gas Analysis and Optimization in Smart Contracts"
By
Mr. Soroush FAROKHNIA
Abstract:
Blockchains are a family of distributed consensus protocols that maintain a
shared, append-only ledger without relying on a trusted central authority.
Smart contracts are self-enforcing programs executed on blockchains that can
support arbitrarily complex logic. Their applications span multiple domains,
including finance, healthcare, and supply chain management, in both the public
and private sectors. They are currently in charge of billions of dollars'
worth of digital assets. On modern programmable blockchains, users incur
execution fees, known as "gas," when interacting with contracts. In the last
five years, Ethereum users spent more than 19.6 billion USD on gas fees. The
high cost of gas and the existence of "out-of-gas" vulnerabilities, which have
led to millions of dollars in user losses, make the analysis and minimization
of gas usage a critical research problem. Yet, existing approaches often
produce suboptimal results, lack strong assurances, or do not scale. These
limitations frequently compel developers to rely on manual audits, which are
costly, error-prone, and can leave vulnerabilities undiscovered.
The main body of this thesis explores the impact of gas from multiple
perspectives. It first approaches the topic from the user's point of view,
focusing on gas estimation and its subsequent reduction through improved
compiler optimizations. It then shifts to the miner's perspective, analyzing
strategies for revenue maximization on two major cryptocurrencies, Ethereum
and Cardano. The primary objective is to develop scalable solutions that
provide strong formal assurances. To this end, this work leverages techniques
such as parameterized algorithms and algebro-geometric methods, which are
particularly effective at exploiting the structural sparsity inherent in smart
contracts. Furthermore, it illustrates that blockchain research is
fundamentally interdisciplinary, utilizing methodologies beyond traditional
computer science to uncover costly vulnerabilities. Finally, each core
contribution is backed by comprehensive empirical evaluations, demonstrating
significant improvements over state-of-the-art solutions.
Date: Thursday, 26 March 2026
Time: 4:00pm - 6:00pm
Venue: Room 2132C
Lift 22
Chairman: Prof. Hai YANG (CIVL)
Committee Members: Dr. Jiasi SHEN (Supervisor)
Dr. Amir GOHARSHADY (Co-supervisor, Oxford)
Prof. Ke YI
Prof. Charles ZHANG
Dr. Maximilian NITZSCHNER (MATH)
Dr. Guillermo PÉREZ (University of Antwerp)