Table of Contents
Fetching ...

Bodyless Block Propagation: TPS Fully Scalable Blockchain with Pre-Validation

Chonghe Zhao, Shengli Zhang, Taotao Wang, Soung Chang Liew

TL;DR

This work tackles the TPS bottleneck in public blockchains by introducing Bodyless Block Propagation (BBP), which propagates only blockheaders while enabling peers to pre-pack and pre-validate the upcoming blockbody. By coordinating pre-packed blockbody (PPB) generation via Time-specific transaction Selection and Ordering (TSO) and synchronizing PPBs across nodes, BBP achieves constant-like block propagation time independent of block size, theoretically yielding $TPS = O(1)$ as compared to the $O(1/\log(N))$ bound for traditional schemes. The authors validate the approach with Ethereum-focused measurements and a large-scale testbed, showing a roughly 4x reduction in block propagation time relative to Ethereum, and demonstrate that maximum per-node TPS can reach around 2335 under favorable settings, while maintaining security through pre-validation and MEV-aware considerations. However, robustness against certain adversarial PPB manipulations (Attacks II/III) requires additional defenses, such as GossipSub-based neighbor scoring. Overall, BBP offers a practical, modular path to full TPS scalability without sacrificing decentralization or security, and integrates as an extension to existing blockchain architectures.

Abstract

Despite numerous prior attempts to boost transaction per second (TPS) of blockchain systems, many sacrifice decentralization and security. This paper proposes a bodyless block propagation (BBP) scheme for which the blockbody is not validated and transmitted during block propagation, to increase TPS without compromising security. Nodes in the blockchain network anticipate the transactions and their ordering in the next upcoming block so that these transactions can be pre-executed and pre-validated before the block is born. For a network with $N$ nodes, our theoretical analysis reveals that BBP can improve TPS scalability from $O(1/log(N))$ to $O(1)$. Ensuring consensus on the next block's transaction content is crucial. We propose a transaction selection, ordering, and synchronization algorithm to drive this consensus. To address the undetermined Coinbase address issue, we further present an algorithm for such unresolvable transactions, ensuring a consistent and TPS-efficient scheme. With BBP, most transactions require neither validation nor transmission during block propagation, liberating system from transaction-block dependencies and rendering TPS scalable. Both theoretical analysis and experiments underscore BBP's potential for full TPS scalability. Experimental results reveal a 4x reduction in block propagation time compared to Ethereum blockchain, with TPS performance being limited by node hardware rather than block propagation.

Bodyless Block Propagation: TPS Fully Scalable Blockchain with Pre-Validation

TL;DR

This work tackles the TPS bottleneck in public blockchains by introducing Bodyless Block Propagation (BBP), which propagates only blockheaders while enabling peers to pre-pack and pre-validate the upcoming blockbody. By coordinating pre-packed blockbody (PPB) generation via Time-specific transaction Selection and Ordering (TSO) and synchronizing PPBs across nodes, BBP achieves constant-like block propagation time independent of block size, theoretically yielding as compared to the bound for traditional schemes. The authors validate the approach with Ethereum-focused measurements and a large-scale testbed, showing a roughly 4x reduction in block propagation time relative to Ethereum, and demonstrate that maximum per-node TPS can reach around 2335 under favorable settings, while maintaining security through pre-validation and MEV-aware considerations. However, robustness against certain adversarial PPB manipulations (Attacks II/III) requires additional defenses, such as GossipSub-based neighbor scoring. Overall, BBP offers a practical, modular path to full TPS scalability without sacrificing decentralization or security, and integrates as an extension to existing blockchain architectures.

Abstract

Despite numerous prior attempts to boost transaction per second (TPS) of blockchain systems, many sacrifice decentralization and security. This paper proposes a bodyless block propagation (BBP) scheme for which the blockbody is not validated and transmitted during block propagation, to increase TPS without compromising security. Nodes in the blockchain network anticipate the transactions and their ordering in the next upcoming block so that these transactions can be pre-executed and pre-validated before the block is born. For a network with nodes, our theoretical analysis reveals that BBP can improve TPS scalability from to . Ensuring consensus on the next block's transaction content is crucial. We propose a transaction selection, ordering, and synchronization algorithm to drive this consensus. To address the undetermined Coinbase address issue, we further present an algorithm for such unresolvable transactions, ensuring a consistent and TPS-efficient scheme. With BBP, most transactions require neither validation nor transmission during block propagation, liberating system from transaction-block dependencies and rendering TPS scalable. Both theoretical analysis and experiments underscore BBP's potential for full TPS scalability. Experimental results reveal a 4x reduction in block propagation time compared to Ethereum blockchain, with TPS performance being limited by node hardware rather than block propagation.
Paper Structure (30 sections, 6 theorems, 7 equations, 11 figures, 1 table, 2 algorithms)

This paper contains 30 sections, 6 theorems, 7 equations, 11 figures, 1 table, 2 algorithms.

Key Result

Lemma 2.1

Broadcasting one transaction to $1-\epsilon$ portion of the nodes can be finished in $O(log(N)+log(1/\epsilon))$ time units.

Figures (11)

  • Figure 1: Illustration of block propagations: (a) classical block propagation; (b) bodyless block propagation.
  • Figure 2: InVestigation Results in Ethereum Blockchain: (a) Block Validation Time; (b) Matched-Transaction and Matched-Blockbody Rates.
  • Figure 3: System architecture of BBP by extending TxPool.
  • Figure 4: Execution for un-intersected transactions.
  • Figure 5: The nodes distribution and expected link delays in our experiments.
  • ...and 6 more figures

Theorems & Definitions (8)

  • Lemma 2.1
  • Lemma 2.2
  • Theorem 2.3
  • Lemma 2.4
  • proof
  • Lemma 2.5
  • proof
  • Theorem 2.6