Supporting Increment and Decrement Operations in Balancing Networks

Aiello, William; Busch, Costas; Herlihy, Maurice; Mavronicolas, Marios; Shavit, Nir; Touitou, Dan

doi:10.1007/3-540-49116-3_37

Cited by 11 publications

(43 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Elimination is a technique introduced by Shavit and Touitou [20] to achieve scalability in shared pool and counter implementations [2,20]. A recent paper by Hendler et.…”

Section: Eliminationmentioning

confidence: 99%

“…Elimination is a parallelization technique that has shown promise in designing scalable shared counters [2,20] and Last-In-First-Out (LIFO) structures such as pools and stacks [7,20]. This paper shows the first example of applying elimination to First-In-First-Out (FIFO) structures, specifically, to one of the most fundamental and widely studied concurrent data structures in the literature: the concurrent FIFO queue [6,9,10,12,14,17,18,19,22,23,24,25].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Using elimination to implement scalable and lock-free FIFO queues

Moir

Nussbaum

Shalev

et al. 2005

Proceedings of the Seventeenth Annual ACM Symposium on Parallelism in Algorithms and Architectures

Self Cite

View full text Add to dashboard Cite

This paper shows for the first time that elimination, a scaling technique formerly applied only to counters and LIFO structures, can be applied to FIFO data structures, specifically, to linearizable FIFO queues. We show how to transform existing nonscalable FIFO queue implementations into scalable implementations using the elimination technique, while preserving lock-freedom and linearizablity.We apply our transformation to the FIFO queue algorithm of Michael and Scott, which is included in the Java TM Concurrency Package. Empirical evaluation on a state-ofthe-art CMT multiprocessor chip shows that by using elimination as a backoff technique for the Michael and Scott queue algorithm, we can achieve comparable performance at low loads, and improved scalability as load increases.

show abstract

“…Elimination is a technique introduced by Shavit and Touitou [20] to achieve scalability in shared pool and counter implementations [2,20]. A recent paper by Hendler et.…”

Section: Eliminationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Using elimination to implement scalable and lock-free FIFO queues

Moir

Nussbaum

Shalev

et al. 2005

Proceedings of the Seventeenth Annual ACM Symposium on Parallelism in Algorithms and Architectures

Self Cite

View full text Add to dashboard Cite

show abstract

“…This necessary condition simply says that all these properties must satisfy the characterization and therefore are closed under the nullity of a balancing network. Consequently, from the results of [2], we can infer that the the step property, the K-smoothing property, and in general the boundedness property are all closed under the nullity of a balancing network. Furthermore, from the results of [5], we can infer that the threshold property is closed under the nullity of a balancing network.…”

Section: Introductionmentioning

confidence: 96%

“…Subsequently, Aiello et al [2] generalized the results of Shavit and Touitou [16] to far more general classes of balancing networks and properties of balancing networks. More specifically, Aiello et al considered boundedness properties, a generalization of the step and K-smoothing properties.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Combinatorial Characterization of Properties Preserved by Antitokens

Busch

Demetriou

Herlihy

et al. 2000

Euro-Par 2000 Parallel Processing

Self Cite

View full text Add to dashboard Cite

Abstract. Balancing networks are highly distributed data structures used to solve multiprocessor synchronization problems. Typically, balancing networks are accessed by tokens, and the distribution of the tokens on the network's output specify the property of the network. However, tokens represent increment operations only, and tokens alone are not adequate for synchronization problems that require decrement operations. For such kinds of problems, antitokens have been used to represent decrement operations. It has been shown that several kinds of balancing networks which satisfy the step property, smoothing property, and the threshold property for tokens alone, preserve their properties even when antitokens are introduced. A fundamental question that was left open was to characterize all the properties of balancing networks which are preserved under the introduction of antitokens. In this work, we provide such a simple combinatorial characterization for all the properties which are preserved when antitokens are introduced.

show abstract

Adding Networks

Fatourou

Herlihy

2001

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Supporting Increment and Decrement Operations in Balancing Networks

Cited by 11 publications

References 15 publications

Using elimination to implement scalable and lock-free FIFO queues

Using elimination to implement scalable and lock-free FIFO queues

A Combinatorial Characterization of Properties Preserved by Antitokens

Adding Networks

Contact Info

Product

Resources

About