[Paper Review] Improve Blockchain Performance using Graph Data Structure and Parallel Mining

Jia Kan, Shangzhe Chen+1 more|arXiv (Cornell University)|Aug 31, 2018

Blockchain Technology Applications and Security4 citations

TL;DR

This paper proposes GraphChain, a blockchain architecture that replaces the traditional chain data structure with a directed acyclic graph (DAG) to enable parallel mining and improve transaction throughput. By allowing multiple miners to validate transactions simultaneously and using a graph-based structure to manage dependencies, the system achieves higher performance and scalability, with simulations showing significant gains in TPS and reduced confirmation times.

ABSTRACT

Blockchain technology is ushering in another break-out year, the challenge of blockchain still remains to be solved. This paper analyzes the features of Bitcoin and Bitcoin-NG system based on blockchain, proposes an improved method of implementing blockchain systems by replacing the structure of the original chain with the graph data structure. It was named GraphChain. Each block represents a transaction and contains the balance status of the traders. Additionally, as everyone knows all the transactions in Bitcoin system will be baled by only one miner that will result in a lot of wasted effort, so another way to improve resource utilization is to change the original way to compete for miner to election and parallel mining. Researchers simulated blockchain with graph structure and parallel mining through python, and suggested the conceptual new graph model which can improve both capacity and performance.

Motivation & Objective

• Address the scalability and performance bottlenecks in existing blockchain systems like Bitcoin and Ethereum, which are limited by sequential mining and low Transactions Per Second (TPS).
• Overcome the inefficiency of Proof of Work (PoW) where most mining efforts are wasted due to single-winner competition.
• Improve network stability and availability by enabling multiple miners to operate in parallel, reducing the risk of network congestion.
• Design a new blockchain framework that supports high-performance applications such as microtransactions, big data storage, and next-generation banking systems.

Proposed method

• Replace the traditional linear blockchain structure with a graph data structure (GraphChain), where each block represents a single transaction and maintains balance states of involved parties.
• Implement a parallel mining mechanism allowing multiple miners to validate and add transactions simultaneously, eliminating the sequential bottleneck of single-leader mining.
• Use a leader election protocol to coordinate mining activities, enabling multiple leaders to operate in parallel while maintaining consistency and avoiding forks.
• Design a storage partitioning strategy to distribute transaction data across nodes, improving scalability and reducing full node storage overhead.
• Simulate the GraphChain system using Python to evaluate performance under varying numbers of miners and transaction loads.
• Integrate the graph structure with PoW consensus to retain security and decentralization while enhancing throughput.

Experimental results

Research questions

• RQ1 Can replacing the linear blockchain structure with a graph-based data structure significantly improve transaction throughput and reduce confirmation latency?
• RQ2 To what extent does parallel mining—allowing multiple miners to validate transactions simultaneously—enhance blockchain performance compared to traditional sequential mining?
• RQ3 How does the GraphChain architecture maintain consistency and prevent forks when multiple miners operate in parallel?
• RQ4 What performance improvements can be achieved in terms of TPS and system availability when scaling the number of miners in the GraphChain model?
• RQ5 Can GraphChain support high-throughput use cases such as microtransactions, decentralized data storage, and financial systems, given its architectural changes?

Key findings

• The GraphChain model significantly increases transaction throughput (TPS) by enabling parallel mining, with performance improving as the number of miners increases, though not linearly.
• Transaction confirmation time is reduced because multiple miners can process transactions simultaneously, eliminating the need to wait for a single block to be mined.
• The system demonstrates improved network stability and availability, as the probability of all nodes going offline decreases with more miners participating.
• The graph-based data structure allows for efficient data partitioning and shard-like handling of transactions, making it suitable for storing large-scale data on the blockchain.
• Simulations show that increasing the number of miners leads to higher system capacity, with performance gains observed even under high transaction loads.
• GraphChain supports practical use cases such as microtransactions and decentralized data systems by enabling fast, low-cost transaction processing without relying on PoS or DPoS.