GLU: A high-level system for granular data-parallel programming

Jagannathan, R.; Dodd, Chris; Agi, I.

doi:10.1002/(sici)1096-9128(199701)9:1<63::aid-cpe240>3.0.co;2-v

Cited by 16 publications

(15 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides the GLU (Granular Lucid) parallel programming system [15] the application of dataflow programming to parallel architectures never gained momentum. Now, 20 years later, there is a veritable renaissance in parallel architectures with multicore general purpose CPUs abounding, and many-core machines now a commodity through GPUs.…”

Section: Discussionmentioning

confidence: 99%

Ypnos

Orchard

Bolingbroke

Mycroft

2010

Proceedings of the 5th ACM SIGPLAN Workshop on Declarative Aspects of Multicore Programming

View full text Add to dashboard Cite

A fully automatic, compiler-driven approach to parallelisation can result in unpredictable time and space costs for compiled code. On the other hand, a fully manual approach to parallelisation can be long, tedious, prone to errors, hard to debug, and often architecturespecific. We present a declarative domain-specific language, Ypnos, for expressing structured grid computations which encourages manual specification of causally sequential operations but then allows a simple, predictable, static analysis to generate optimised, parallel implementations. We introduce the language and provide some discussion on the theoretical aspects of the language semantics, particularly the structuring of computations around the category theoretic notion of a comonad.

show abstract

Section: Discussionmentioning

confidence: 99%

Ypnos

Orchard

Bolingbroke

Mycroft

2010

Proceedings of the 5th ACM SIGPLAN Workshop on Declarative Aspects of Multicore Programming

View full text Add to dashboard Cite

show abstract

“…The identifiers whose evaluation depends on them can then be evaluated, and eventually the initial identifier is evaluated [26]. This demand-driven eduction model naturally supports distributed execution of intensional programs and has been used in various solutions before [14,25].…”

Section: Gipsy and Dmfmentioning

confidence: 99%

Unifying and refactoring DMF to support concurrent Jini and JMS DMS in GIPSY

Mokhov

Paquet

2012

Proceedings of the Fifth International C* Conference on Computer Science and Software Engineering

View full text Add to dashboard Cite

The General Intensional Programming System (GIPSY) is a framework for the compilation and distributed demanddriven evaluation of context-aware declarative programs. Its distributed run-time system includes a demand migration framework that has up to now been instantiated with two different communication technologies, namely Jini and JMS. However, the different nature, APIs, and requirements of these two solutions have resulted in artifact implementation divergence from the original demand migration framework (DMF) specifications as well as flaws in the DMF itself. This also inhibited the concurrent consistent use of the technologies within the same GIPSY network instance.Thus, in this paper we report on our re-engineering effort and results to refactor and unify the two somewhat disjoint Java distributed middleware technologies -Jini and JMSused in the implementation of the corresponding Demand Migration Systems (DMS). In doing so, we refactor their parent Demand Migration Framework (DMF), within the General Intensional Programming System (GIPSY) and realign the Jini and JMS implementation remedying the flaws, improving interoperability, and allowing for scalability testing with various real applications and comparative studies.

show abstract

“…GLU [7,6], JLucid, and later Objective Lucid [12] prompted the development of a General Imperative Compiler Framework (GICF). The framework targets the integration different imperative languages into GIPSY programs for I/O, portability, extensibility, and flexibility reasons.…”

Section: General Intensional Programming System (Gipsy)mentioning

confidence: 99%

Towards Security Hardening of Scientific Demand-Driven and Pipelined Distributed Computing Systems

Mokhov

2008

2008 International Symposium on Parallel and Distributed Computing

View full text Add to dashboard Cite

This work highlights and takes aim at the most critical security aspects required for two different types of distributed systems for scientific computation. It covers two open-source systems written in Java: a demand-driven system -General Intensional Programming System (GIPSY) and a pipelined system -Distributed Modular Audio Recognition Framework (DMARF), which are the distributed scientific computational engines used as case studies with respect to the security aspects. More specific goals include data/demand integrity, data/demand origin authentication, confidentiality, high availability, and malicious code detection. We address some of the goals to a degree, some with the Java Data Security Framework (JDSF) as a work-inprogress.

show abstract

GLU: A high-level system for granular data-parallel programming

Cited by 16 publications

References 15 publications

Ypnos

Ypnos

Unifying and refactoring DMF to support concurrent Jini and JMS DMS in GIPSY

Towards Security Hardening of Scientific Demand-Driven and Pipelined Distributed Computing Systems

Contact Info

Product

Resources

About