Using elimination to implement scalable and lock-free FIFO queues

Moir, Mark; Nussbaum, Daniel A.; Shalev, Ori; Shavit, Nir

doi:10.1145/1073970.1074013

Cited by 89 publications

(65 citation statements)

References 22 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We run up to 64 threads on the 32 cores. The influence of garbage collection was negligible for all algorithm 8 .…”

Section: Experiments Setupmentioning

confidence: 99%

“…Another approach for reducing contention is using elimination, as proposed by Moir et al [8]. Here, producers and consumers can "eliminate" each other at predefined rendezvous points.…”

Section: Related Workmentioning

confidence: 99%

“…Producer/consumer elimination. We have also adopted the elimination technique used in [8] and [1]. Consider a thread t that inserted a new task at a node, and started updating the node's ancestors.…”

Section: Treecontainermentioning

confidence: 99%

See 2 more Smart Citations

CAFÉ: Scalable Task Pools with Adjustable Fairness and Contention

Basin

Fan

Keidar

et al. 2011

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Task pools have many important applications in distributed and parallel computing. Pools are typically implemented using concurrent queues, which limits their scalability. We introduce CAFÉ, Contention and Fairness Explorer, a scalable and wait-free task pool which allows users to control the trade-off between fairness and contention. The main idea behind CAFÉ is to maintain a list of TreeContainers, a novel tree-based data structure providing efficient task inserts and retrievals. TreeContainers don't guarantee FIFO ordering on task retrievals. But by varying the size of the trees, CAFÉ can provide any type of pool, from ones using large trees with low contention but less fairness, to ones using small trees with higher contention but also greater fairness. We demonstrate the scalability of TreeContainer by proving an O(log 2 N ) bound on the step complexity of insert operations when there are N inserts, as compared to an average of Ω(N ) steps in a queue based implementation. We further prove that get operations are wait-free. Evaluations of CAFÉ show that it outperforms the Java SDK implementation of the Michael-Scott queue by a factor of 30, and is over three times faster than other state-of-the-art non-FIFO task pools.

show abstract

“…We run up to 64 threads on the 32 cores. The influence of garbage collection was negligible for all algorithm 8 .…”

Section: Experiments Setupmentioning

confidence: 99%

“…Another approach for reducing contention is using elimination, as proposed by Moir et al [8]. Here, producers and consumers can "eliminate" each other at predefined rendezvous points.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

CAFÉ: Scalable Task Pools with Adjustable Fairness and Contention

Basin

Fan

Keidar

et al. 2011

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…The minimum of these two values is stored to send (statement 15). This number of responses is now sent in the direction specified by the dir field of the first requests entry (statements 16,17). Finally, send responses and requests are removed from the n.resp and n.pending queues, respectively.…”

Section: The Combining Processmentioning

confidence: 99%

“…Their empirical results show that their algorithm achieves high parallelism in practice; nevertheless, it does not provide any deterministic guarantee of parallelism. Moir et al used ideas similar to these of [10] to obtain a queue algorithm that possesses the same properties [17].…”

Section: Shavit and Zemach Introduce Diffracting Treesmentioning

confidence: 99%

Constructing Shared Objects That Are Both Robust and High-Throughput

Hendler

Kutten

2006

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Shared counters are among the most basic coordination structures in distributed computing. Known implementations of shared counters are either blocking, non-linearizable, or have a sequential bottleneck. We present the first counter algorithm that is both linearizable, nonblocking, and can provably achieve high throughput in semisynchronous executions. The algorithm is based on a novel variation of the software combining paradigm that we call bounded-wait combining. It can thus be used to obtain implementations, possessing the same properties, of any object that supports combinable operations, such as stack or queue. Unlike previous combining algorithms where processes may have to wait for each other indefinitely, in the bounded-wait combining algorithm a process only waits for other processes for a bounded period of time and then 'takes destiny in its own hands'. In order to reason rigorously about the parallelism attainable by our algorithm, we define a novel metric for measuring the throughput of shared objects which we believe is interesting in its own right. We use this metric to prove that our algorithm can achieve throughput of Ω(N/ log N ) in executions where process speeds vary only by a constant factor, where N is the number of processes that can participate in the algorithm. We also introduce and use pseduo-transactions -a technique for concurrent execution that may prove useful for other algorithms.

show abstract

Lock‐Free Concurrent Data Structures

Cederman

Gidenstam

et al. 2017

Programming Multi‐core and Many‐core Computing Systems

View full text Add to dashboard Cite

Using elimination to implement scalable and lock-free FIFO queues

Cited by 89 publications

References 22 publications

CAFÉ: Scalable Task Pools with Adjustable Fairness and Contention

CAFÉ: Scalable Task Pools with Adjustable Fairness and Contention

Constructing Shared Objects That Are Both Robust and High-Throughput

Lock‐Free Concurrent Data Structures

Contact Info

Product

Resources

About