Exploiting coarse-grained task, data, and pipeline parallelism in stream programs

Gordon, Michael I.; Thies, William; Amarasinghe, Saman

doi:10.1145/1168857.1168877

Cited by 349 publications

(154 citation statements)

References 32 publications

Supporting

Mentioning

142

Contrasting

Order By: Relevance

“…Finally, we are heavily looking into parallel computing techniques to improve the simulation of quantum operations, at the moment carried out by straightforward linear algebraic techniques. Concretely, we are investigating the compilation of command sequences into a dataflow network (Gordon et al, 2006). In such a network quantum states are represented by a long stream of amplitudes, which has the double benefit of exposing the inherent parallelism while at the same time relaxing the need to fit entire vectors inside the same computer memory.…”

Section: Resultsmentioning

confidence: 99%

Distributed Quantum Programming

D’Hondt

Vandriessche

2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

In this paper we explore the structure and applicability of the Distributed Measurement Calculus (DMC), an assembly language for distributed measurement-based quantum computations. We describe the formal language's syntax and semantics, both operational and denotational, and state several properties that are crucial to the practical usability of our language, such as equivalence of our semantics, as well as compositionality and context-freeness of DMC programs. We show how to put these properties to use by constructing a composite program that implements distributed controlled operations, in the knowledge that the semantics of this program does not change under the various composition operations. Our formal model is the basis of a quantum virtual machine construction for distributed quantum computations, which we elaborate upon in the latter part of this work. This virtual machine embodies the formal semantics of DMC such that programming execution no longer needs to be analysed by hand. Far from a literal translation, it requires a substantial concretisation of the formal model at the level of data structures, naming conventions and abstraction mechanisms. At the same time we provide automatisation techniques for program specification where possible to obtain an expressive and user-friendly programming environment.

show abstract

Section: Resultsmentioning

confidence: 99%

Distributed Quantum Programming

D’Hondt

Vandriessche

2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…It has been compiled on a myriad of architectures including multi-cores [9], Cell [17], Raw [8] and even FPGAs [10]. Moreover, there is previous research in coarse-grained mapping of StreamIt to GPU platforms [23].…”

Section: Related Workmentioning

confidence: 99%

Scalable framework for mapping streaming applications onto multi-GPU systems

Huynh

Hagiescu

Wong

et al. 2012

Proceedings of the 17th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming

View full text Add to dashboard Cite

Graphics processing units leverage on a large array of parallel processing cores to boost the performance of a specific streaming computation pattern frequently found in graphics applications. Unfortunately, while many other general purpose applications do exhibit the required streaming behavior, they also possess unfavorable data layout and poor computation-to-communication ratios that penalize any straight-forward execution on the GPU. In this paper we describe an efficient and scalable code generation framework that can map general purpose streaming applications onto a multi-GPU system. This framework spans the entire core and memory hierarchy exposed by the multi-GPU system. Several key features in our framework ensure the scalability required by complex streaming applications. First, we propose an efficient stream graph partitioning algorithm that partitions the complex application to achieve the best performance under a given shared memory constraint. Next, the resulting partitions are mapped to multiple GPUs using an efficient architecture-driven strategy. The mapping balances the workload while considering the communication overhead. Finally, a highly effective pipeline execution is employed for the execution of the partitions on the multi-GPU system. The framework has been implemented as a back-end of the StreamIt programming language compiler. Our comprehensive experiments show its scalability and significant performance speedup compared with a previous stateof-the-art solution.

show abstract

“…range (1,10) here is a logical constant and a Java expression, denoting a set of all integers from 1 through 10. By contrast, y == x * x is a PQL subexpression: both x and y are logical variables.…”

Section: Language Constructs Overviewmentioning

confidence: 99%

PQL: A Purely-Declarative Java Extension for Parallel Programming

Reichenbach

Smaragdakis

Immerman

2012

ECOOP 2012 – Object-Oriented Programming

View full text Add to dashboard Cite

Abstract. The popularization of parallelism is arguably the most fundamental computing challenge for years to come. We present an approach where parallel programming takes place in a restricted (sub-Turing-complete), logic-based declarative language, embedded in Java. Our logic-based language, PQL, can express the parallel elements of a computing task, while regular Java code captures sequential elements. This approach offers a key property: the purely declarative nature of our language allows for aggressive optimization, in much the same way that relational queries are optimized by a database engine. At the same time, declarative queries can operate on plain Java data, extending patterns such as map-reduce to arbitrary levels of nesting and composition complexity. We have implemented PQL as extension to a Java compiler and showcase its expressiveness as well as its scalability compared to competitive techniques for similar tasks (Java + relational queries, in-memory Hadoop, etc.).

show abstract

Exploiting coarse-grained task, data, and pipeline parallelism in stream programs

Cited by 349 publications

References 32 publications

Distributed Quantum Programming

Distributed Quantum Programming

Scalable framework for mapping streaming applications onto multi-GPU systems

PQL: A Purely-Declarative Java Extension for Parallel Programming

Contact Info

Product

Resources

About