Scalable lock-free FIFO queues using efficient elimination techniques

Authors:
V. V. N. Pavan Kumar;K. Gopinath
Affiliations:
Advanced Systems Laboratory (ASL), DRDO, Hyderabad;Indian Institute of Science, Bangalore
Venue:
Proceedings of the 5th ACM COMPUTE Conference: Intelligent & scalable system technologies
Year:
2012

Citing 6
Cited 0

Diffracting trees

ACM Transactions on Computer Systems (TOCS)
Simple, fast, and practical non-blocking and blocking concurrent queue algorithms

PODC '96 Proceedings of the fifteenth annual ACM symposium on Principles of distributed computing
Using elimination to implement scalable and lock-free FIFO queues

Proceedings of the seventeenth annual ACM symposium on Parallelism in algorithms and architectures
Scalable synchronous queues

Proceedings of the eleventh ACM SIGPLAN symposium on Principles and practice of parallel programming
A scalable lock-free stack algorithm

Journal of Parallel and Distributed Computing
Supporting increment and decrement operations in balancing networks

STACS'99 Proceedings of the 16th annual conference on Theoretical aspects of computer science

Quantified Score

Hi-index	0.00

Visualization

Abstract

Lock-free data structures ensure system wide progress, synchronizing the operations of concurrent processes without the use of mutual exclusion. Lock-free FIFO queues are one of the most highly studied concurrent data structures. Elimination techniques have been used to improve the scalability of FIFO queues. In this paper, we study scalable lock-free FIFO queues using newer and more efficient elimination techniques. We propose four new heuristics to improves the scalability of the Lock-Free FIFO queues: Range Variation, Fixed Location, Load Adaptation and Double Array. Our empirical results shows these four heuristics perform better than existing algorithms in the literature.