Componentising a Scientific Application for the Grid

Abstract

Building and evolving Grid applications is complex. Component-based development has emerged as an effective approach to building flexible systems, but there is little experience in applying this approach to Grid programming. This paper presents our experience with reengineering a high performance numerical solver to become a component-based Grid application. The adopted component model is an extension of the generic Fractal model that specifically targets grid environments. The paper provides qualitative and quantitative evidence that componentisation has improved the modifiability and reusability of the application while not significantly affecting performance.

Collective Interfaces for Distributed Components

“…We already used a prototype of the ProActive/GCM to build and deploy a numerical computation application for electromagnetism, on 4 clusters and more than 300 processors [19]. Although this prototype did not use collective interfaces, this experimentation allowed us to show the scalability of the model.…”

Performance and Scalability of a Component-Based Grid Application

“…Figure 2 shows the main activities and artefacts defined by the componentisation process. The activities are summarised next (more details can be found in [18] .…”

COMP Superscalar: Bringing GRID Superscalar and GCM Together

“…In this sense, following the methodology explained in [12], we studied the architecture of the GRIDSs runtime in order to identify its main functionalities (namely dependency analysis, scheduling, file managing and job handling) and then we assigned each of them to a separate component. As a result, the runtime has gained in reusability, deployability, flexibility and separation of concerns; moreover, developing and improving it is simpler now, since new features can be added to one of the components without having to modify the others.…”