Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
140 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

FastPay: High-Performance Byzantine Fault Tolerant Settlement (2003.11506v3)

Published 25 Mar 2020 in cs.CR

Abstract: FastPay allows a set of distributed authorities, some of which are Byzantine, to maintain a high-integrity and availability settlement system for pre-funded payments. It can be used to settle payments in a native unit of value (crypto-currency), or as a financial side-infrastructure to support retail payments in fiat currencies. FastPay is based on Byzantine Consistent Broadcast as its core primitive, foregoing the expenses of full atomic commit channels (consensus). The resulting system has low-latency for both confirmation and payment finality. Remarkably, each authority can be sharded across many machines to allow unbounded horizontal scalability. Our experiments demonstrate intra-continental confirmation latency of less than 100ms, making FastPay applicable to point of sale payments. In laboratory environments, we achieve over 80,000 transactions per second with 20 authorities---surpassing the requirements of current retail card payment networks, while significantly increasing their robustness.

Citations (43)

Summary

  • The paper introduces FastPay, a novel BFT-based RTGS system that leverages Byzantine consistent broadcast to minimize overhead and ensure rapid transaction finality.
  • The paper demonstrates experimental results showing over 80,000 transactions per second and sub-100ms latency using a sharded authority design under Byzantine faults.
  • The paper reveals significant implications for retail and crypto payments by eliminating interbank risks and enabling scalable, decentralized settlement.

Overview of FastPay: High-Performance Byzantine Fault Tolerant Settlement

The paper introduces FastPay, a Byzantine Fault Tolerant (BFT) real-time gross settlement (RTGS) system designed to deliver high-integrity and availability for pre-funded payments. Unlike traditional settlement systems, FastPay uses Byzantine Consistent Broadcast to avoid the overhead and complexity of full atomic commits, achieving low-latency transaction finality. FastPay supports both crypto and fiat payments with a sharded architecture for unlimited horizontal scalability, reaching transaction rates that surpass existing retail card networks.

FastPay's design breaks from conventional RTGS and blockchain models by eliminating interbank settlement risks and enhancing the robustness often associated with centralized financial systems. The primary contribution of FastPay is its ability to maintain safety, liveness, and high performance under Byzantine failures, supported by formal proofs. The system tolerates Byzantine failures by relying on $3f+1$ authorities, with operational resilience ensured as long as ff authorities can be Byzantine out of the total.

Key Contributions and Results

  • Design Innovation: By foregoing full consensus, FastPay minimizes shared state between accounts, leveraging the innate properties of payments for increased concurrency while retaining efficiency through sharded authorities.
  • Scale and Performance: The experimental evaluation shows FastPay achieving over 80,000 transactions per second with 20 authorities, exhibiting intra-continental confirmation latency of less than 100ms. This performance is maintained despite high concurrency and Byzantine faults, significantly exceeding the requirements of the busiest retail card networks.
  • Applicability: FastPay's sub-second latency and robustness make it ideal for retail and point-of-sale payments, while its horizontal scalability ensures suitability for high-demand environments. It can act both as a standalone settlement solution, a side-chain for cryptocurrencies, or an auxiliary layer to traditional RTGS.

Theoretical and Practical Implications

FastPay presents a transformative approach to settling retail payments by dismissing the need for intermediating banks, consequently negating associated credit and counterparty risks. It necessitates rethinking the architecture of BFT real-time settlement systems, moving from traditional centralized designs to more open, scalable, and resilient frameworks.

In theoretical terms, FastPay aligns with recent advances in distributed systems that pursue high efficiency without consensus complexities. It demonstrates a pragmatic application of Byzantine consistent broadcast techniques in financial contexts, positioning it as both a viable competitor to and complement of existing blockchain-based solutions and RTGS systems.

Speculation on Future Developments

Successful implementation and scaling of FastPay could prompt a paradigm shift in high-frequency trading and retail payment systems, pushing further developments in BFT mechanisms and decentralized finance infrastructures. We can anticipate enhanced collaboration between legacy systems and decentralized architectures, sparking new innovations in financial technology ecosystems.

The formal grounding and experimental validation of FastPay propose a benchmark for future endeavors to address existing shortcomings in speed and reliability of settlement systems. Continued exploration into integrating privacy mechanisms, governance structures, and fee architectures can bolster FastPay's capabilities and expand its real-world applicability, mitigating current limitations and enhancing its adoption potential.

In conclusion, FastPay exemplifies a significant advancement in BFT settlement systems by optimizing for performance, scalability, and fault tolerance, setting an emergent standard for future research and application in high-performance financial networks.

Github Logo Streamline Icon: https://streamlinehq.com
Youtube Logo Streamline Icon: https://streamlinehq.com