Bitcoin-NG: A Scalable Blockchain Protocol (1510.02037v2)

Abstract: Cryptocurrencies, based on and led by Bitcoin, have shown promise as infrastructure for pseudonymous online payments, cheap remittance, trustless digital asset exchange, and smart contracts. However, Bitcoin-derived blockchain protocols have inherent scalability limits that trade-off between throughput and latency and withhold the realization of this potential. This paper presents Bitcoin-NG, a new blockchain protocol designed to scale. Based on Bitcoin's blockchain protocol, Bitcoin-NG is Byzantine fault tolerant, is robust to extreme churn, and shares the same trust model obviating qualitative changes to the ecosystem. In addition to Bitcoin-NG, we introduce several novel metrics of interest in quantifying the security and efficiency of Bitcoin-like blockchain protocols. We implement Bitcoin-NG and perform large-scale experiments at 15% the size of the operational Bitcoin system, using unchanged clients of both protocols. These experiments demonstrate that Bitcoin-NG scales optimally, with bandwidth limited only by the capacity of the individual nodes and latency limited only by the propagation time of the network.

• The paper presents Bitcoin-NG, a protocol that separates leader election from transaction processing to significantly reduce consensus delay and enhance scalability.
• It introduces key blocks for leader selection and microblocks for rapid transaction serialization, ensuring efficient and continuous ledger updates.
• Experimental evaluations on a 1000-node network show improved fairness, mining power utilization, and optimal throughput compared to Bitcoin.

An Analysis of the Bitcoin-NG Protocol

"Bitcoin-NG: A Scalable Blockchain Protocol," authored by Ittay Eyal, Adem Efe Gencer, Emin Gün Sirer, and Robbert van Renesse, presents a novel blockchain protocol aimed at scaling Bitcoin's infrastructure to overcome its inherent throughput and latency limitations. This essay provides a detailed summary and examination of the key contributions, methodologies, and experimental evaluations of Bitcoin-NG, emphasizing its significance for future developments in blockchain technology.

Introduction and Background

Bitcoin's introduction of a decentralized ledger via the Nakamoto consensus protocol has enabled a variety of applications, from decentralized currency to smart contracts. However, Bitcoin’s scalability remains a core issue, restricted by the interplay between block size and block interval. Increasing the block size delays network propagation, whereas decreasing the block interval risks system instability, making it necessary to trade off between throughput and latency. Bitcoin-NG addresses these constraints by adopting a new protocol design that separates leader election from the transaction serialization process.

Design of Bitcoin-NG

Bitcoin-NG modifies Bitcoin’s protocol by introducing two distinct types of blocks: key blocks and microblocks. Key blocks are generated via a proof-of-work mechanism, similar to Bitcoin, and serve to elect a leader. Each elected leader then has the authority to serialize transactions in microblocks which are produced without substantial proof-of-work.

Key Innovations:

1. Leader Election and Epochs: Bitcoin-NG divides time into epochs, with each epoch designated to a single leader elected via key blocks. Key blocks are generated at an exponential interval, akin to Bitcoin’s current block generation.
2. Transaction Serialization: The elected leader generates microblocks at a high frequency, containing ledger entries and transactions. Microblocks facilitate continuous transaction processing, constrained only by network propagation time and node processing capabilities.
3. Fork Handling and Remuneration: While Bitcoin-NG key blocks carry weight similar to Bitcoin blocks, microblocks do not influence chain weight, thus mitigating selfish mining vulnerabilities. Remuneration in the form of transaction fees is split between the current leader and the subsequent leader to prevent transaction withholding. Furthermore, fork prevention is handled via poison transactions which invalidate revenue for leaders that cause forks.

Metrics and Experimental Setup

The paper introduces several metrics to quantify the performance and security of blockchain protocols, including consensus delay, fairness, mining power utilization, and time to prune. Evaluations were conducted on a 1000-node emulated network using modified Bitcoin clients, ensuring conditions closely mirrored those of an operational Bitcoin network.

Experimental Results

1. Consensus Delay and Time to Prune: Bitcoin-NG significantly outperforms Bitcoin in terms of consensus delay and time to prune. Despite the higher microblock frequency, consensus delays remain contingent solely on network propagation time.
2. Fairness and Mining Power Utilization: Bitcoin-NG maintains optimal fairness and mining power utilization across all tested conditions, unlike Bitcoin, which shows significant performance degradation at high block frequencies and sizes. Specifically, Bitcoin sees a marked reduction in fairness and mining power utilization as block frequency increases, leading to potential security risks due to increased fork rates.
3. Throughput and Scalability: The throughput of Bitcoin-NG is limited only by individual node capacity, achieving high transaction frequencies without compromising security metrics. The evaluations demonstrated that Bitcoin-NG can scale optimally, supporting higher throughput without encountering Bitcoin's bandwidth and latency trade-offs.

Implications and Future Directions

Bitcoin-NG offers a robust solution to the scalability issues faced by Bitcoin, maintaining security and performance under higher transaction loads. By decoupling leader election from transaction serialization, Bitcoin-NG can achieve low latency and high throughput. The methodologies and metrics introduced set a foundation for evaluating and comparing alternative consensus protocols, providing a concrete path forward for blockchain scalability research.

The implications of Bitcoin-NG are substantial for the development of decentralized applications, potentially enabling use cases that require higher transaction rates and lower latency. Future work may explore combining Bitcoin-NG with other protocols like GHOST or Inclusive Blockchains to enhance performance further or address practical challenges.

In conclusion, Bitcoin-NG represents a significant advancement in the design of scalable blockchain protocols, addressing the critical bottlenecks of Bitcoin's current implementation. The findings and innovations of this paper will likely influence the trajectory of blockchain research and development, opening avenues for more efficient and scalable decentralized systems.