Algorithmic skeletons meeting grids

Danelutto, Marco; Aldinucci, Marco

doi:10.1016/j.parco.2006.04.001

Cited by 23 publications

(10 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ASSIST [26,3], AutoMate [23], K-Components [14], SAFRAN [13] and finally the forthcoming CoreGRID Component Model (GCM) [11] all include autonomic management features. The latter two are derived from a common ancestor, i.e.…”

Section: Related Workmentioning

confidence: 99%

“…It is currently under development by the partners of the EU CoreGRID Network of Excellence 1 . A companion EU STREP project, GridCOMP 2 is currently developing an open source implementation of GCM and preliminary versions are already available for download as embedded modules in the ProActive middleware suite 3 . GCM builds on the Fractal component model [21] and exhibits three prominent features: hierarchical composition, collective interactions and autonomic management.…”

Section: Gcm: the Grid Component Modelmentioning

confidence: 99%

“…Also, the stateless functional replication idea can be extended to the stateful case by requiring inner components Ws to expose suitable methods to serialise, read and write the internal state. A suitable manipulation of the serialised state enables the reconfiguration of workers (also in the data-parallel scenario [3]). …”

Section: Behavioural Skeletonsmentioning

confidence: 99%

“…As a consequence, grid applications need to dynamically adapt to the features of the underlying architecture in order to be efficient and/or high performance [3].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Behavioural Skeletons in GCM: Autonomic Management of Grid Components

Aldinucci

Campa

Danelutto

et al. 2008

16th Euromicro Conference on Parallel, Distributed and Network-Based Processing (PDP 2008)

Self Cite

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Section: Gcm: the Grid Component Modelmentioning

confidence: 99%

Section: Behavioural Skeletonsmentioning

confidence: 99%

“…As a consequence, grid applications need to dynamically adapt to the features of the underlying architecture in order to be efficient and/or high performance [3].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Behavioural Skeletons in GCM: Autonomic Management of Grid Components

Aldinucci

Campa

Danelutto

et al. 2008

16th Euromicro Conference on Parallel, Distributed and Network-Based Processing (PDP 2008)

Self Cite

View full text Add to dashboard Cite

“…Another type, which can be useful for more distributed and loosely-coupled scenarios, appears as component collections , as in the XCAT framework (Govindaraju et al, 2003) or the ProActive implementation of the Fractal component model (Baduel et al, 2006) with Grid Component Model (GCM) (Baude et al, 2009) extensions, including collective interfaces (Baude et al, 2007). The skeleton approach can be used, as in ASSIST (Danelutto and Aldinucci, 2006) and HOC-SA (Dünnweber and Gorlatch, 2009). The choice of the underlying programming model can restrict the high-level composition types available for applications.…”

Section: State Of the Artmentioning

confidence: 99%

Component-based approach for programming and running scientific applications on grids and clouds

Malawski

Gubała

Bubak

2011

The International Journal of High Performance Computing Applica

View full text Add to dashboard Cite

This paper presents an approach to programming and running scientific applications on grid and cloud infrastructures based on two principles: the first one is to follow a component-based programming model, the second is to apply a flexible technology which allows for virtualization of the underlying infrastructure. The solutions described in this paper include high-level composition and deployment consisting of a scripting-based environment and a manager system based on an architecture description language (ADL), a dynamically managed pool of component containers, and interoperability with other component models such as Grid Component Model (GCM). We demonstrate how the proposed methodology can be implemented by combining the unique features of the Common Component Architecture (CCA) model together with the H2O resource sharing platform, resulting in the MOCCA component framework. Applications and tests include data mining using the Weka library, Monte Carlo simulation of the formation of clusters of gold atoms, as well as a set of synthetic benchmarks. The conclusion is that the component approach to scientific applications can be successfully applied to both grid and cloud infrastructures.

show abstract

Fault tolerance for data parallel programs

Bertolli

Vanneschi

2010

Concurrency and Computation

View full text Add to dashboard Cite

The main issues when supporting fault tolerance based on checkpointing and rollback recovery for High-Performance applications are related to the scalability of the introduced support, the possibility of analyzing the induced overhead and, in more general terms, the optimization of the trade-off between failure-free and recovery performances. In this paper we describe our contribution in fault tolerance for high-level structured parallelism models. We take a different viewpoint w.r.t. existing contributions, by introducing a methodology to derive interesting properties to support fault tolerance. We show how to apply this methodology to a general data parallel model, deriving useful properties to introduce a class of checkpointing protocols. Thanks to this methodology, this class of protocols is not affected by the described issues. We exemplify two checkpointing protocols and the related rollback recovery techniques. For each protocol we also derive cost models statically describing the failure-free performance, which can be used for performance tuning or to target some Quality of Service parameter. To assess the innovation of the results we analytically and experimentally compare the introduced protocols with two literature protocols. Results show that while the protocols introduced in this paper permit the definition of cost models and have a good scalability, the literature protocols do not always have these properties

show abstract

Algorithmic skeletons meeting grids

Cited by 23 publications

References 12 publications

Behavioural Skeletons in GCM: Autonomic Management of Grid Components

Behavioural Skeletons in GCM: Autonomic Management of Grid Components

Component-based approach for programming and running scientific applications on grids and clouds

Fault tolerance for data parallel programs

Contact Info

Product

Resources

About