SymGrid-Par: Designing a Framework for Executing Computational Algebra Systems on Computational Grids

Zain, Abdallah Al; Hammond, Kevin; Trinder, Phil; Linton, Steve; Loidl, Hans-Wolfgang; Costanti, Marco

doi:10.1007/978-3-540-72586-2_90

Cited by 10 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SymGridPar2 (SGP2) is middleware for coordinating large‐scale task‐parallel computations distributed over many networked GAP instances. SymGridPar2 inherits its architecture, and specifically, its skeleton‐based programming model, from SymGridPar . The key new feature is the use of HdpH (Haskell distributed parallel Haskell), a novel domain‐specific language (DSL) for task‐parallel programming on large‐scale networks, including HPC platforms.…”

Section: The Design and Implementation Of Symgridpar2mentioning

confidence: 99%

HPC‐GAP: engineering a 21st‐century high‐performance computer algebra system

Behrends

Hammond

Janjić

et al. 2016

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

SUMMARYSymbolic computation has underpinned a number of key advances in Mathematics and Computer Science. Applications are typically large and potentially highly parallel, making them good candidates for parallel execution at a variety of scales from multi-core to high-performance computing systems. However, much existing work on parallel computing is based around numeric rather than symbolic computations. In particular, symbolic computing presents particular problems in terms of varying granularity and irregular task sizes that do not match conventional approaches to parallelisation. It also presents problems in terms of the structure of the algorithms and data. This paper describes a new implementation of the free open-source GAP computational algebra system that places parallelism at the heart of the design, dealing with the key scalability and cross-platform portability problems. We provide three system layers that deal with the three most important classes of hardware: individual shared memory multi-core nodes, mid-scale distributed clusters of (multi-core) nodes and full-blown high-performance computing systems, comprising large-scale tightly connected networks of multi-core nodes. This requires us to develop new cross-layer programming abstractions in the form of new domain-specific skeletons that allow us to seamlessly target different hardware levels. Our results show that, using our approach, we can achieve good scalability and speedups for two realistic exemplars, on high-performance systems comprising up to 32 000 cores, as well as on ubiquitous multi-core systems and distributed clusters. The work reported here paves the way towards full-scale exploitation of symbolic computation by high-performance computing systems, and we demonstrate the potential with two major case studies.

show abstract

Section: The Design and Implementation Of Symgridpar2mentioning

confidence: 99%

HPC‐GAP: engineering a 21st‐century high‐performance computer algebra system

Behrends

Hammond

Janjić

et al. 2016

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

show abstract

“…This effort attempts to go the other way around and "gridify" applications such as Matlab and Octave, in order for them to be able to exploit distributed resources. This is also done for Maple in [15] and for a variety of computer algebra packages in [18]. While these works are significant, it is questionable how the produced services can be used in an integrated service oriented environment during the service lifecycle.…”

Section: Related Workmentioning

confidence: 99%

A Service-Oriented Framework for GNU Octave-Based Performance Prediction

Kousiouris

Kyriazis

Konstanteli

et al. 2010

2010 IEEE International Conference on Services Computing

View full text Add to dashboard Cite

Cloud/Grid environments are characterized by a diverse set of technologies used for communication, execution and management. Service Providers, in this context, need to be equipped with an automated process in order to optimize service provisioning through advanced performance prediction methods. Furthermore, existing software solutions such as GNU Octave offer a wide range of possibilities for implementing these methods. However, their automated use as services in the distributed computing paradigm includes a number of challenges from a design and implementation point of view. In this paper, a loosely coupled service-oriented implementation is presented, for taking advantage of software like Octave in the process of creating and using prediction models during the service lifecycle of a SOI. In this framework, every method is applied as an Octave script in a plug-in fashion. The design and implementation of the approach is validated through a case study application which involves the transcoding of raw video to MPEG4.

show abstract

“…Figure 1 depicts the SymGrid-Par design as a stack of layers (left) of increasing levels of abstraction. The middle stack presents an early version of SymGrid-Par, based on a bespoke interface between Haskell and the underlying CAS [1]. The right stack describes the latest, standards-compliant version of SymGrid-Par, supporting a distributed collection of servers.…”

Section: Introductionmentioning

confidence: 99%

SymGrid-Par

project¹

2011

ACM Commun. Comput. Algebra

View full text Add to dashboard Cite

SymGrid-Par: Designing a Framework for Executing Computational Algebra Systems on Computational Grids

Cited by 10 publications

References 14 publications

HPC‐GAP: engineering a 21st‐century high‐performance computer algebra system

HPC‐GAP: engineering a 21st‐century high‐performance computer algebra system

A Service-Oriented Framework for GNU Octave-Based Performance Prediction

SymGrid-Par

Contact Info

Product

Resources

About