PIPQ: Strict Insert-Optimized Concurrent Priority Queue

2025-08-21

AI Summary

This work addresses the performance bottleneck of concurrent priority queues under insertion-intensive workloads. We propose the first strictly linearizable priority queue explicitly designed to maximize insertion parallelism—departing from conventional approaches that prioritize delete-min optimization. Our design employs a two-tier architecture: a lock-free, thread-local worker layer for high-throughput insertions, and a leader layer that maintains a compact global heap of high-priority elements to ensure low-latency delete-min operations. Fine-grained synchronization and ordered merge protocols guarantee strict linearizability while maximizing insertion throughput. Experimental evaluation demonstrates that our queue outperforms state-of-the-art alternatives by up to 3.2× under insertion-dominated workloads, and achieves superior scalability and consistency guarantees in real-world applications such as graph processing.

Technology Category

• Search and Optimization: Distributed Search
• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Security and Privacy: Large-scale security measurements

Abstract

This paper presents PIPQ, a strict and linearizable concurrent priority queue whose design differs from existing solutions in literature because it focuses on enabling parallelism of insert operations as opposed to accelerating delete-min operations, as traditionally done. In a nutshell, PIPQ's structure includes two levels: the worker level and the leader level. The worker level provides per-thread data structures enabling fast and parallel insertions. The leader level contains the highest priority elements in the priority queue and can thus serve delete-min operations. Our evaluation, which includes an exploration of different data access patterns, operation mixes, runtime settings, and an integration into a graph-based application, shows that PIPQ outperforms competitors in a variety of cases, especially with insert-dominant workloads.

Problem

• Optimizing concurrent insert operations in priority queues
• Enabling parallelism for insertions over delete-min operations
• Designing a two-level structure for efficient concurrent access

Innovation

• Two-level structure for parallel insertions
• Worker level enables fast concurrent inserts
• Leader level handles delete-min operations efficiently