Grevm

Grevm (both 1.0 and 2.0) is reth-ready, please see use-with-reth.md for more details.

Grevm is a Block-STM inspired optimistic parallel EVM execution engine that leverages DAG-based task scheduling, dynamic dependency management, and parallel state storage to significantly boost throughput of revm, while reducing CPU overhead in high-conflict scenarios.

TL;DR – Highlights of Grevm 2.0

• Grevm 2.0 achieves near-optimal performance in low-contention scenarios, matching Block-STM with 16.57 gigagas/s for Uniswap workloads and outperforming it with 95% less CPU usage in inherently non-parallelizable cases by 20–30%, achieving performance close to sequential execution.
• Breaks Grevm 1.0’s limitations in handling highly dependent transactions, delivering a 5.5× throughput increase to 2.96 gigagas/s in 30%-hot-ratio hybrid workloads by minimizing re-executions through DAG-based scheduling and Task Groups.
• Introduces Parallel State Store, leveraging asynchronous execution result bundling to overlap and amortize 30-60ms of post-execution overhead within parallel execution, effectively hiding these costs within execution time. It also seamlessly handles miner rewards and the self-destruct opcode without the performance penalties of sequential fallbacks.
• In-depth analysis of optimistic parallel execution reveals the underestimated efficiency of Block-STM and the strength of optimistic parallelism, providing new insights into parallel execution.

Architecture Overview

Grevm 2.0 is composed of three main modules:

• Dependency Manager (DAG Manager):
  Constructs a directed acyclic graph (DAG) of transaction dependencies based on speculative read/write hints.

• Execution Scheduler:
  Selects transactions with no dependencies (out-degree of 0) for parallel execution, groups adjacent dependent transactions into task groups, and dynamically updates dependencies to minimize re-execution.

• Parallel State Storage:
  Provides an asynchronous commit mechanism with multi-version memory to reduce latency and manage miner rewards and self-destruct opcodes efficiently.

Running the Benchmark

To reproduce the benchmark:

JEMALLOC_SYS_WITH_MALLOC_CONF="thp:always,metadata_thp:always" \
NUM_EOA=<num_accounts> HOT_RATIO=<hot_ratio> DB_LATENCY_US=<latency_in_us> \
cargo bench --bench gigagas

Replace <num_accounts>, <hot_ratio>, and <latency_in_us> with your desired parameters.

Further Details

For a comprehensive explanation of the design, algorithmic choices, and in-depth benchmark analysis, please refer to the full technical report.

• Grevm 2.0 Tech Report
• Grevm 1.0 Tech Report