Semantics-aware scheduling policies for synchronization determinism

Zhao, Qi; Qiu, Zhengyi; Jin, Guoliang

doi:10.1145/3293883.3295731

Cited by 5 publications

(3 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use the most recent release (October-2018 for Kepler, and September-2019 for Pegasus) of the state-of-the-art HPC workflow managers. These are the most widely used workflow managers in the HPC community [38,69,75,88], although these approaches are agnostic to serverless computing platforms. Given a workflow DAG and a VM cluster, they perform a profiling run to learn the structure and introduce several optimizations to decide how to schedule different applications to exploit parallelism.…”

Section: Experimental Methodologymentioning

confidence: 99%

Mashup

Roy

Patel

Gadepally

et al. 2022

Proceedings of the 27th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming

View full text Add to dashboard Cite

This work introduces Mashup, a novel strategy to leverage serverless computing model for executing scientific workflows in a hybrid fashion by taking advantage of both the traditional VM-based cloud computing platform and the emerging serverless platform. Mashup outperforms the state-ofthe-art workflow execution engines by an average of 34% and 43% in terms of execution time reduction and cost reduction, respectively, for widely-used HPC workflows on the Amazon Cloud platform (EC2 and Lambda). CCS Concepts: • Computer systems organization → Cloud computing; Heterogeneous (hybrid) systems; • Computing methodologies → Distributed computing methodologies; Massively parallel and high-performance simulations.

show abstract

Section: Experimental Methodologymentioning

confidence: 99%

Mashup

Roy

Patel

Gadepally

et al. 2022

Proceedings of the 27th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming

View full text Add to dashboard Cite

show abstract

“…Moreover, it characterizes races between events that can be arbitrarily far apart in the input trace, as opposed to happens-before and other partial-order methods that only characterize races between successive conflicting accesses. Our notion is applicable to all concurrency settings, and interestingly, it is also complete for systems with synchronizationdeterministic concurrency [Aguado et al 2018;Bocchino et al 2009;Cui et al 2015;Zhao et al 2019]. 2We develop an efficient, single-pass, nearly linear time algorithm SyncP that, given a trace , detects whether contains a sync-preserving race.…”

Section: Introductionmentioning

confidence: 99%

Optimal prediction of synchronization-preserving races

Mathur

Pavlogiannis

Viswanathan

2021

Proc. ACM Program. Lang.

View full text Add to dashboard Cite

Concurrent programs are notoriously hard to write correctly, as scheduling nondeterminism introduces subtle errors that are both hard to detect and to reproduce. The most common concurrency errors are (data) races, which occur when memory-conflicting actions are executed concurrently. Consequently, considerable effort has been made towards developing efficient techniques for race detection. The most common approach is dynamic race prediction: given an observed, race-free trace σ of a concurrent program, the task is to decide whether events of σ can be correctly reordered to a trace σ * that witnesses a race hidden in σ. In this work we introduce the notion of sync(hronization)-preserving races. A sync-preserving race occurs in σ when there is a witness σ * in which synchronization operations (e.g., acquisition and release of locks) appear in the same order as in σ. This is a broad definition that strictly subsumes the famous notion of happens-before races. Our main results are as follows. First, we develop a sound and complete algorithm for predicting sync-preserving races. For moderate values of parameters like the number of threads, the algorithm runs in Õ( N ) time and space, where N is the length of the trace σ. Second, we show that the problem has a Ω( N /log 2 N ) space lower bound, and thus our algorithm is essentially time and space optimal. Third, we show that predicting races with even just a single reversal of two sync operations is NP-complete and even W1-hard when parameterized by the number of threads. Thus, sync-preservation characterizes exactly the tractability boundary of race prediction, and our algorithm is nearly optimal for the tractable side. Our experiments show that our algorithm is fast in practice, while sync-preservation characterizes races often missed by state-of-the-art methods.

show abstract

“…Moreover, it characterizes races between events that can be arbitrarily far apart in the input trace, as opposed to happens-before and other partial-order methods that only characterize races between successive conflicting accesses. Our notion is applicable to all concurrency settings, and interestingly, it is also complete for systems with synchronization-deterministic concurrency [Aguado et al 2018;Bocchino et al 2009;Cui et al 2015;Zhao et al 2019].…”

Section: Introductionmentioning

confidence: 99%

Optimal Prediction of Synchronization-Preserving Races

Mathur¹,

Pavlogiannis²,

Viswanathan³

2020

Preprint

View full text Add to dashboard Cite

Concurrent programs are notoriously hard to write correctly, as scheduling nondeterminism introduces subtle errors that are both hard to detect and to reproduce. The most common concurrency errors are (data) races, which occur when memory-conflicting actions are executed concurrently. Consequently, considerable effort has been made towards developing efficient techniques for race detection. The most common approach is dynamic race prediction: given an observed, race-free trace 𝜎 of a concurrent program, the task is to decide whether events of 𝜎 can be correctly reordered to a trace 𝜎 * that witnesses a race hidden in 𝜎.In this work we introduce the notion of sync(hronization)-preserving races. A sync-preserving race occurs in 𝜎 when there is a witness 𝜎 * in which synchronization operations (e.g., acquisition and release of locks) appear in the same order as in 𝜎. This is a broad definition that strictly subsumes the famous notion of happens-before races. Our main results are as follows. First, we develop a sound and complete algorithm for predicting sync-preserving races. For moderate values of parameters like the number of threads, the algorithm runs in 𝑂 (N ) time and space, where N is the length of the trace 𝜎. Second, we show that the problem has a Ω(N /log 2 N ) space lower bound, and thus our algorithm is essentially time and space optimal. Third, we show that predicting races with even just a single reversal of two sync operations is NP-complete and even W[1]-hard when parameterized by the number of threads. Thus, sync-preservation characterizes exactly the tractability boundary of race prediction, and our algorithm is nearly optimal for the tractable side. Our experiments show that our algorithm is fast in practice, while sync-preservation characterizes races often missed by state-of-the-art methods.CCS Concepts: • Software and its engineering → Software verification and validation; • Theory of computation → Theory and algorithms for application domains; Program analysis.

show abstract

Semantics-aware scheduling policies for synchronization determinism

Cited by 5 publications

References 44 publications

Mashup

Mashup

Optimal prediction of synchronization-preserving races

Optimal Prediction of Synchronization-Preserving Races

Contact Info

Product

Resources

About