Litcius/Paper detail

Evaluating the Efficiency of zk-SNARK, zk-STARK, and Bulletproof in Real-World Scenarios: A Benchmark Study

Mohammed El‐Hajj, Bjorn Oude Roelink

2024Information17 citationsDOIOpen Access PDF

Abstract

This study builds on our previous systematic literature review (SLR) that assessed the applications and performance of zk-SNARK, zk-STARK, and Bulletproof non-interactive zero-knowledge proof (NIZKP) protocols. To address the identified research gaps, we designed and implemented a benchmark comparing these three protocols using a dynamic minimized multiplicative complexity (MiMC) hash application. We evaluated performance across four general-purpose programming libraries and two programming languages. Our results show that zk-SNARK produced the smallest proofs, while zk-STARK generated the largest. In terms of proof generation and verification times, zk-STARK was the fastest, and Bulletproof was the slowest. Interestingly, zk-SNARK proofs verified marginally faster than zk-STARK, contrary to other findings. These insights enhance our understanding of the functionality, security, and performance of NIZKP protocols, providing valuable guidance for selecting the most suitable protocol for specific applications.

Topics & Concepts

Benchmark (surveying)Mathematical proofMultiplicative functionComputer scienceProtocol (science)Theoretical computer scienceMathematicsGeometryMathematical analysisMedicineGeographyPathologyAlternative medicineGeodesySecurity and Verification in ComputingForensic Toxicology and Drug AnalysisDistributed systems and fault tolerance
Evaluating the Efficiency of zk-SNARK, zk-STARK, and Bulletproof in Real-World Scenarios: A Benchmark Study | Litcius