Online Block Packing

2025-07-16

AI Summary

This paper studies the online block packing problem for “quasi-patient” bidders under multi-dimensional block resource constraints. Addressing the absence of efficient approximation algorithms for this setting, we propose the first online approximation algorithm with provable theoretical guarantees. Our approach jointly models multi-dimensional resource capacity constraints and bidders’ bounded waiting times (i.e., quasi-patient behavior), integrating threshold-based pricing with dynamic bin-packing techniques to achieve a constant-factor approximation ratio in polynomial time. The algorithm resolves an open problem posed at EC 2025. It significantly improves the trade-off among throughput, resource utilization, and fairness in blockchain systems, offering both a theoretically sound foundation and a practically deployable solution for dynamic block packing in real-world blockchain consensus protocols.

Technology Category

• Multiagent Systems: Mechanism Design
• Machine Learning: Online Learning & Bandits
• Game Theory and Economic Paradigms: Mechanism Design

Application Category

• Economics, Online Markets and Human Computation: Economic aspects of blockchain and cryptocurrencies
• Security and Privacy: Blockchains and distributed ledgers
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

We consider the algorithmic challenge that is faced by blockchains that have multidimensional block constraints and serve quasi-patient bidders. We provide online approximation algorithms for this problem, thus solving open problems left by [Babaioff and Nisan, EC 2025].

Problem

• Address multidimensional block constraints in blockchains
• Serve quasi-patient bidders in blockchain systems
• Provide online approximation algorithms for block packing

Innovation

• Online algorithms for multidimensional block constraints
• Approximation solutions for blockchain challenges
• Addressing open problems from prior research