Xiaofei Bai (School of Computer Science, Fudan University), Jian Gao (School of Computer Science, Fudan University), Chenglong Hu (School of Computer Science, Fudan University), Liang Zhang (School of Computer Science, Fudan University)

Blockchain networks, especially cryptocurrencies, rely heavily on proof-of-work (PoW) systems, often as a basis to distribute rewards. These systems require solving specific puzzles, where Application Specific Integrated Circuits (ASICs) can be designed for performance or efficiency. Either way, ASICs surpass CPUs and GPUs by orders of magnitude, and may harm blockchain networks. Recently, Equihash is developed to resist ASIC solving with heavy memory usage. Although commercial ASIC solvers exist for its most popular parameter set, such solvers do not work under better ones, and are considered impossible under optimal parameters. In this paper, we inspect the ASIC resistance of Equihash by constructing a parameter-independent adversary solver design. We evaluate the product, and project at least 10x efficiency advantage for resourceful adversaries. We contribute to the security community in two ways: (1) by revealing the limitation of Equihash and raising awareness about its algorithmic factors, and (2) by demonstrating that security inspection is practical and useful on PoW systems, serving as a start point for future research and development.

View More Papers

Giving State to the Stateless: Augmenting Trustworthy Computation with...

Gabriel Kaptchuk (Johns Hopkins University), Matthew Green (Johns Hopkins University), Ian Miers (Cornell Tech)

Read More

NIC: Detecting Adversarial Samples with Neural Network Invariant Checking

Shiqing Ma (Purdue University), Yingqi Liu (Purdue University), Guanhong Tao (Purdue University), Wen-Chuan Lee (Purdue University), Xiangyu Zhang (Purdue University)

Read More

SABRE: Protecting Bitcoin against Routing Attacks

Maria Apostolaki (ETH Zurich), Gian Marti (ETH Zurich), Jan Müller (ETH Zurich), Laurent Vanbever (ETH Zurich)

Read More

Neuro-Symbolic Execution: Augmenting Symbolic Execution with Neural Constraints

Shiqi Shen (National University of Singapore), Shweta Shinde (National University of Singapore), Soundarya Ramesh (National University of Singapore), Abhik Roychoudhury (National University of Singapore), Prateek Saxena (National University of Singapore)

Read More