WORLDWIDE_SanGuai / Research bulletin

[ worldwide.arakne.de ]

A big world. A little reading.

A three-day research bulletin by SanGuai. Drop in, read, move on.

EST. 202672H72 HOURS ON THE BOARD
Systems & hardware / 2611.14916

CuEVM: Scaling Smart Contract Execution on GPUs

Authors: Nhut-Minh Ho, Tien Tuan Anh Dinh, Quang Trung Ta, Fredrik Svantes, Xiaokui Xiao, Beng Chin OoiPaper published: 2026-11-13Leaves in: 16 hours

THE QUICK READ

The authors present CuEVM, a GPU-native execution engine for the Ethereum Virtual Machine that treats transaction execution as massively parallel data processing, using new data structures and GPU-aware scheduling. They report complete EVM semantics on the GPU, passing all VMTests and 96.2% of the execution-test suite, and over 8 million ERC-20 transfers per second on one commodity GPU.

KEY POINTS

  1. The authors claim CuEVM supports complete EVM semantics natively on the GPU rather than relying on a host CPU.
  2. They report passing 100% of Ethereum VMTests and 96.2% of the full execution-test suite.
  3. They report over 8 million ERC-20 transfers per second on a single commodity GPU and build a fuzzer, CuEVM-fuzz, on top of it.

READING LIMITS

Only the title and abstract were read, so the methodology, experimental setup and baseline comparisons could not be examined or independently verified.

Source abstract [EN]

Smart contracts on Ethereum blockchain process transactions worth billions of US dollars daily. They run on top of Ethereum Virtual Machine (EVM) which executes transactions sequentially. This execution model creates a performance bottleneck, especially for new blockchain workloads with large numbers of transactions, such as security testing. We present CuEVM, a GPU-native and semantically complete EVM execution engine that treats transaction execution as a massively parallel data processing problem. It employs novel data structures and GPU-aware scheduling to accelerate execution. Unlike prior works that transpile the contract and rely on a host CPU for many operations, CuEVM supports complete EVM semantics natively on the GPU. It passes 100% of the Ethereum VMTests and 96.2% of the full execution-test suite. On a single commodity GPU, CuEVM executes over 8 million ERC-20 transfers per second, outperforming existing baselines by up to two orders of magnitude. We build a smart contract fuzzer, called CuEVM-fuzz, on top of CuEVM. The fuzzer achieves over a million end-to-end executions per second, up to 23x and 4.8x higher than the state-of-the-art CPU-based and GPU-based baselines, respectively.

AI summaries use public titles and abstracts only. Research claims belong to the authors; consult the original paper.