[Paper Review] Chain of Trust: Can Trusted Hardware Help Scaling Blockchains?

Hung Dang, Anh Dinh+2 more|arXiv (Cornell University)|Apr 2, 2018

Blockchain Technology Applications and Security35 references7 citations

TL;DR

This paper proposes leveraging trusted hardware to enhance the scalability of blockchain consensus protocols by introducing two design principles: 'scale up by scaling down' and 'prioritize consensus messages.' These principles enable significant throughput improvements—up to 7x in Hyperledger Fabric and up to 32x reduction in shard creation cost in sharding protocols—while maintaining system operation as networks grow.

ABSTRACT

As blockchain systems proliferate, there remains an unresolved scalability problem of their underlying distributed consensus protocols. Byzantine Fault Tolerance (BFT) consensus protocols offer high transaction throughput, but can only support small networks. Proof-of-Work (PoW) consensus protocol, on the other hand, can support thousands of nodes, but at the expense of poor transaction throughput. Two potential approaches to address these scalability barriers are by relaxing the threat model, or by employing sharding technique to deal with large networks. Nonetheless, their effectiveness on data-intensive blockchain workloads remains to be seen. In this work, we study the use and effectiveness of trusted hardware on scaling distributed consensus protocols, and by their extension, blockchain systems. We first analyze existing approaches that harness trusted hardware to enhance scalability, and identify their limitations. Drawing insights from these results, we propose two design principles, namely scale up by scaling down and prioritize consensus messages, enabling the consensus protocols to scale. We illustrate the two principles by presenting optimizations that improve upon state-of-the-art solutions, and demonstrate via our extensive evaluations that they indeed offer better scalability. In particular, integrating our optimizations into Hyperledger Fabric achieves up to 7x higher throughput, while enabling the system to remain operational as the network expands. Another optimization that we introduce to Hyperledger Sawtooth allows the system to sustain high throughput as the network grows. Finally, our new design for sharding protocols reduces the cost of shard creation phase by up to 32x.

Motivation & Objective

• To address the scalability limitations of blockchain consensus protocols, particularly the trade-off between network size and transaction throughput.
• To investigate whether trusted hardware can effectively enhance scalability in Byzantine Fault Tolerance (BFT) and Proof-of-Work (PoW) consensus protocols.
• To identify limitations in existing trusted hardware-based approaches and derive design principles for improved scalability.
• To optimize existing blockchain systems like Hyperledger Fabric and Sawtooth using these principles, achieving higher throughput and scalability.

Proposed method

• Introduce two design principles: 'scale up by scaling down' to reduce per-node overhead and 'prioritize consensus messages' to improve message processing efficiency.
• Apply these principles to optimize Hyperledger Fabric, reducing consensus message processing latency and increasing throughput.
• Design a trusted hardware-secured sharding protocol that minimizes the cost of shard creation using hardware-enforced integrity and confidentiality.
• Use trusted execution environments (TEEs) to enforce message prioritization and reduce trust in individual nodes during consensus.
• Evaluate the proposed optimizations in real-world deployments across varying network sizes and workloads.
• Compare performance against state-of-the-art solutions to quantify gains in throughput and scalability.

Experimental results

Research questions

• RQ1 Can trusted hardware effectively mitigate the scalability bottleneck in BFT and PoW consensus protocols?
• RQ2 How do design principles like 'scale up by scaling down' and 'prioritize consensus messages' improve consensus efficiency in large-scale blockchains?
• RQ3 To what extent can trusted hardware reduce the overhead of sharding protocols, particularly during the shard creation phase?
• RQ4 What performance gains are achievable in real blockchain systems like Hyperledger Fabric and Sawtooth when applying these hardware-optimized designs?
• RQ5 How does the integration of trusted hardware maintain system availability and security as network size increases?

Key findings

• Integrating the proposed optimizations into Hyperledger Fabric achieved up to 7x higher transaction throughput while supporting larger network scales.
• The optimized design for Hyperledger Sawtooth sustained high throughput even as the network size increased, demonstrating improved horizontal scalability.
• The new sharding protocol design reduced the cost of the shard creation phase by up to 32x compared to existing approaches.
• The use of trusted hardware enabled more efficient message prioritization, reducing consensus latency and improving throughput.
• The proposed design principles effectively balanced security and performance, maintaining system operation under increasing load.
• The evaluation confirmed that trusted hardware can be a key enabler for scaling blockchain systems beyond the limitations of traditional consensus protocols.