English

Ira: Efficient Transaction Replay for Distributed Systems

Distributed, Parallel, and Cluster Computing 2026-04-14 v2 Databases

Abstract

In primary-backup replication, consensus latency is bounded by the time for backup nodes to replay (re-execute) transactions proposed by the primary. In this work, we present Ira, a framework to accelerate backup replay by transmitting compact \emph{hints} alongside transaction batches. Our key insight is that the primary, having already executed transactions, possesses knowledge of future access patterns which is exactly the information needed for optimal replay. We use Ethereum for our case study and present a concrete protocol, Ira-L, within our framework to improve cache management of Ethereum block execution. The primaries implementing Ira-L provide hints that consist of the working set of keys used in an Ethereum block and one byte of metadata per key indicating the table to read from, and backups use these hints for efficient block replay. We evaluated Ira-L against the state-of-the-art Ethereum client reth over two weeks of Ethereum mainnet activity (100,800100,800 blocks containing over 2424 million transactions). Our hints are compact, adding a median of 4747 KB compressed per block (5%\sim5\% of block payload). We observe that the sequential hint generation and block execution imposes a 28.6%28.6\% wall-time overhead on the primary, though the direct cost from hints is 10.9%10.9\% of execution time; all of which can be pipelined and parallelized in production deployments. On the backup side, we observe that Ira-L achieves a median per-block speedup of 25×25\times over baseline reth. With 1616 prefetch threads, aggregate replay time drops from 6.56.5 hours to 1616 minutes (23.6×23.6\times wall-time speedup).

Cite

@article{arxiv.2601.21286,
  title  = {Ira: Efficient Transaction Replay for Distributed Systems},
  author = {Adithya Bhat and Harshal Bhadreshkumar Shah and Mohsen Minaei},
  journal= {arXiv preprint arXiv:2601.21286},
  year   = {2026}
}

Comments

Added a disclaimer

R2 v1 2026-07-01T09:25:03.340Z