Efficient synchronization of multiprocessors with shared memory

Kruskal, Clyde P.; Rudolph, Larry; Snir, Marc

doi:10.1145/10590.10609

Cited by 72 publications

(48 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Let φ be an element of Φ and x a variable. The read-modify-write operation [8], RM W (x, φ), atomically replaces the value of x with φ(x), and returns the prior value of x. Most common synchronization primitives, such as fetch&add, swap, test&set, and compare&swap, can be cast as read-modify-write operations for suitable choices of φ.…”

Section: Discussionmentioning

confidence: 99%

Linearizable counting networks

1996

View full text Add to dashboard Cite

Abstract. An adding network is a distributed data structure that supports a concurrent, lock-free, low-contention implementation of a fetch&add counter; a counting network is an instance of an adding network that supports only fetch&increment. We present a lower bound showing that adding networks have inherently high latency. Any adding network powerful enough to support addition by at least two values a and b, where |a| > |b| > 0, has sequential executions in which each token traverses Ω(n/c) switching elements, where n is the number of concurrent processes, and c is a quantity we call one-shot contention; for a large class of switching networks and for conventional counting networks the one-shot contention is constant. On the contrary, counting networks have O(log n) latency [4,7]. This bound is tight. We present the first concurrent, lock-free, lowcontention networked data structure that supports arbitrary fetch&add operations.

show abstract

Section: Discussionmentioning

confidence: 99%

Linearizable counting networks

1996

View full text Add to dashboard Cite

show abstract

“…The algorithm can be used to provide a linearizable, nonblocking and high-throughput implementation of any combinable operation [8,17]. For the sake of presentation simplicity, however, we describe it in the context of implementing a shared counter.…”

Section: An Overview Of the Bwc Algorithmmentioning

confidence: 99%

“…Bounded-wait combining can be easily adapted to work for any combinable operation [8,17] and can thus be used to implement linearizable stacks and queues that are both nonblocking and highly parallel in k-synchronous executions. We are not aware of any other deterministic stack or queue algorithm to possess these properties.…”

Section: Shavit and Zemach Introduced Diffracting Treesmentioning

confidence: 99%

Bounded-wait combining: constructing robust and high-throughput shared objects

Hendler

Kutten

2009

Distrib. Comput.

View full text Add to dashboard Cite

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 k-synchronous executions -executions in which process speeds vary by at most a constant factor k. 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 a stack or a 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 achieves throughput of Ω(N/ log N ) in k-synchronous executions, where N is the number of processes that can participate in the algorithm.Our algorithm uses two tools that we believe may prove useful for obtaining highly parallel nonblocking implementation of additional objects. The first are "synchronous locks", locks that are respected by processes only in k-synchronous executions and are disregarded otherwise; the second are "pseduo-transactions" -a weakening of regular transactions that allows higher parallelism.

show abstract

“…nonblocking and high-throughput implementation of any combinable operation [6,15]. For the sake of presentation simplicity, however, we describe it in the context of implementing a shared counter, supporting the Fetch&increment operation.…”

Section: An Overview Of the Bwc Algorithmmentioning

confidence: 99%

“…Bounded-wait combining can be adapted to work for any combinable operation [6,15] and can thus be used to implement nonblocking and highly parallel linearizable stacks and queues. We are not aware of any other stack or queue algorithm to possess these properties.…”

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

Efficient synchronization of multiprocessors with shared memory

Cited by 72 publications

References 18 publications

Linearizable counting networks

Linearizable counting networks

Bounded-wait combining: constructing robust and high-throughput shared objects

Constructing Shared Objects That Are Both Robust and High-Throughput

Contact Info

Product

Resources

About