Gamma programming paradigm and heterogeneous computing

Murthy, V. K.; Krishnamurthy, E. V.

doi:10.1109/hicss.1996.495472

Cited by 12 publications

(7 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [7], it is shown how a rule-based multiset programming paradigm close to Gamma [4] may be used as a unifying theme for various models of computation, such as biological , molecular, DNA, etc. The Organic Grid [5] is another effort based on a biologically inspired paradigm.…”

Section: Discussionmentioning

confidence: 99%

Towards chemical coordination for grids

Banâtre

Fradet

Radenac

2006

Proceedings of the 2006 ACM Symposium on Applied Computing

View full text Add to dashboard Cite

In [6], Ian Foster and Karl Kesselman explain that grids need "a rethinking of existing programming models and, most likely, new thinking about novel models". In this work, we investigate a "novel programming model" for grids based on the chemical metaphor.On one side, applications are programmed in an abstract manner describing essentially the chemical coordination between (not necessarily chemical) software components. On the other side, chemical service programs are specifically provided to the Grid run-time system in order to obtain the expected quality of service in terms of efficiency, reliability, security, etc. These service programs can be seen as special coordination programs providing guidelines to the runtime system allowing a better use of resources in order to obtain the expected Quality of Service.

show abstract

Section: Discussionmentioning

confidence: 99%

Towards chemical coordination for grids

Banâtre

Fradet

Radenac

2006

Proceedings of the 2006 ACM Symposium on Applied Computing

View full text Add to dashboard Cite

show abstract

“…3.Parallelism between the actions of an object which sends a request message and the actions of an object that 456 receives a message can be permitted depending upon the context. 4.The nature of internal production rules and actions determine whether an object reacts deterministically, nondeterministically or probabilistically [ 15], [ 17].This enables us to: (i) Assign probabilities for applying the rule (ii) Assign strength to each rule by using a measure of its past success (iii) Introduce a support for each rule by using a measure of its likely relevance to the current situation. The above three factors provide for competition and cooperation among the different rules.The introduction of probabilistic choices in an object system would provide a computational model (such as the genetic algorithm) to simulate evolutionary biological, chemical and physical systems based on intermittent feedback from the environment.…”

Section: Achieving Different Types Of Parallelismmentioning

confidence: 99%

Heterogeneous programming with concurrent objects

Murthy

Krishnamurthy

1997

Proceedings of the 1997 ACM Symposium on Applied Computing - SAC '97

View full text Add to dashboard Cite

Heterogeneous computing refers to using a collection of different types computers and achieve maximal concurrency in solving a problem by identifying each computer as an object.This paper is concerned with the heterogeneous concurrent object programming (COP) based on event driven production systems and transactional execution. The usefulness of transactional approach is explained and followed by an examination of the role of event-driven production sytems in COP.The local and global serializability conditions are derived using object dependency graphs. Also we describe a scheduler that can be embedded in an object to ensure global and local serializability. Finally, we describe a simplified version of COP (a diffusing computation). We also indicate how to implement COP using the MPI.

show abstract

“…This would require the analysis as to how the respective internal and external transactions interfere with each other when they are applied; see [18][19][20][21][22][23][24][25]. In a Mobile environment with many Mobile hosts, the above conflicts are hard to resolve due to disconnection ; hence global consisteny through serialization becomes an issue and we must relax this criterion also [24].…”

Section: Concurrency and Serializabilitymentioning

confidence: 99%