Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Boosting Blockchain Throughput: Parallel EVM Execution with Asynchronous Storage for Reddio

Xiaodong Qi, Xinran Chen, Asiy, Neil Han

TL;DR

Reddio tackles the throughput bottleneck of blockchain systems by combining batch-based parallel EVM execution with an asynchronous, pipelined state database. It introduces direct state reading, asynchronous node retrieval, and a coordinated pipeline to decouple computation from storage I/O while ensuring deterministic serializability. The approach includes a formal correctness discussion, a commit-point strategy for state trie nodes, and robust recovery mechanisms to maintain consistency after failures. The results indicate substantial throughput improvements by mitigating storage bottlenecks without sacrificing Ethereum compatibility or security guarantees.

Abstract

The increasing adoption of blockchain technology has led to a growing demand for higher transaction throughput. Traditional blockchain platforms, such as Ethereum, execute transactions sequentially within each block, limiting scalability. Parallel execution has been proposed to enhance performance, but existing approaches either impose strict dependency annotations, rely on conservative static analysis, or suffer from high contention due to inefficient state management. Moreover, even when transaction execution is parallelized at the upper layer, storage operations remain a bottleneck due to sequential state access and I/O amplification. In this paper, we propose Reddio, a batch-based parallel transaction execution framework with asynchronous storage. Reddio processes transactions in parallel while addressing the storage bottleneck through three key techniques: (i) direct state reading, which enables efficient state access without traversing the Merkle Patricia Trie (MPT); (ii) asynchronous parallel node loading, which preloads trie nodes concurrently with execution to reduce I/O overhead; and (iii) pipelined workflow, which decouples execution, state reading, and storage updates into overlapping phases to maximize hardware utilization.

Boosting Blockchain Throughput: Parallel EVM Execution with Asynchronous Storage for Reddio

TL;DR

Reddio tackles the throughput bottleneck of blockchain systems by combining batch-based parallel EVM execution with an asynchronous, pipelined state database. It introduces direct state reading, asynchronous node retrieval, and a coordinated pipeline to decouple computation from storage I/O while ensuring deterministic serializability. The approach includes a formal correctness discussion, a commit-point strategy for state trie nodes, and robust recovery mechanisms to maintain consistency after failures. The results indicate substantial throughput improvements by mitigating storage bottlenecks without sacrificing Ethereum compatibility or security guarantees.

Abstract

The increasing adoption of blockchain technology has led to a growing demand for higher transaction throughput. Traditional blockchain platforms, such as Ethereum, execute transactions sequentially within each block, limiting scalability. Parallel execution has been proposed to enhance performance, but existing approaches either impose strict dependency annotations, rely on conservative static analysis, or suffer from high contention due to inefficient state management. Moreover, even when transaction execution is parallelized at the upper layer, storage operations remain a bottleneck due to sequential state access and I/O amplification. In this paper, we propose Reddio, a batch-based parallel transaction execution framework with asynchronous storage. Reddio processes transactions in parallel while addressing the storage bottleneck through three key techniques: (i) direct state reading, which enables efficient state access without traversing the Merkle Patricia Trie (MPT); (ii) asynchronous parallel node loading, which preloads trie nodes concurrently with execution to reduce I/O overhead; and (iii) pipelined workflow, which decouples execution, state reading, and storage updates into overlapping phases to maximize hardware utilization.

Paper Structure

This paper contains 31 sections, 5 theorems, 4 equations, 7 figures, 6 algorithms.

Table of Contents

  1. Introduction
  2. Organization
  3. Background and Motivation
  4. Blockchain and Smart Contracts
  5. Blockchain
  6. Smart Contracts
  7. Account Model and Contract States
  8. State Database
  9. Merkle Patricia Trie (MPT)
  10. State Database
  11. Workflow
  12. Smart Contract Parallel Execution
  13. Motivation
  14. Overview
  15. Architecture
  16. ...and 16 more sections

Key Result

Lemma 1

The state database $S$ always reflects the effects of executing transactions $\langle T_0, \ldots, T_{I_{\text{next}} - 1} \rangle$ serially after the parallel execution of each batch of transactions.

Figures (7)

  • Figure 1: An example of Merkle Patrica Trie (MPT).
  • Figure 2: An example of the relationship between the account trie and storage trie.
  • Figure 3: Synchronous workflow in the Merkle Patricia Trie.
  • Figure 4: Framework of parallel EVM execution.
  • Figure 5: Asynchronous pipeline of state database in Reddio.
  • ...and 2 more figures

Theorems & Definitions (12)

  • Definition 1: Deterministic Serializability
  • Definition 2: Transaction Conflict
  • Lemma 1
  • proof
  • Lemma 2
  • proof
  • Lemma 3
  • proof
  • Lemma 4
  • proof
  • ...and 2 more