With the advent of Grid and application technologies, scientists and engineers are building more and more complex applications to manage and process large data sets, and execute scientific experiments on distributed resources. Such application scenarios require means for composing and executing complex workflows. Therefore, many efforts have been made towards the development of workflow management systems for Grid computing. In this paper, we propose a taxonomy that characterizes and classifies various approaches for building and executing workflows on Grids. We also survey several representative Grid workflow systems developed by various projects world-wide to demonstrate the comprehensiveness of the taxonomy. The taxonomy not only highlights the design and engineering similarities and differences of state-of-the-art in Grid workflow systems, but also identifies the areas that need further research.

Abstract not found

Parameter-sweep has been widely adopted in large numbers of scientific applications. Parameter-sweep features need to be incorporated into Grid workflows so as to increase the scale and scope of such applications. New scheduling mechanisms and algorithms are required to provide optimized policy for resource allocation and task arrangement in such a case. This paper addresses scheduling sequential parameter-sweep tasks in a fine-grained manner. The optimization is produced by pipelining the subtasks and dispatching each of them onto well-selected resources. Two types of scheduling algorithms are discussed and customized to adapt the characteristics of parameter-sweep, as well as their effectiveness has been compared under multifarious scenarios.

In this paper we propose a framework dedicated to the development and the execution of parallel applications over large scale global computing platforms. A workflow programming environment will be introduced, based on a new workflow language YvetteML and a Human-GRID middleware interface called YML. This language allows description of different kind of components to be allocated to GRID resources. Depending of the different targeted resources, the components may be associated to computation, migration of data or other resource controls. YML is designed to have several back-ends for different middleware, as a welldesigned front end is developed independently of any dedicated middleware. In order to make the framework immediately useful, YML comes with pre-configured interfaces to some numerical routines and a numerical library for iterative linear algebra methods. We will present experimentations done on some large scale platforms using a peer to peer middleware with a numerical application case study. 1.

Abstract. We present a generic and light-weight Grid workflow execution engine made available as a Grid service. A long-term goal is to facilitate the rapid development of application-oriented end-user workflow tools, while providing a high degree of Grid middleware-independence. The workflow engine is designed for workflow execution, independent of client tools for workflow definition. A flexible plugin-structure for middleware-integration provides a strict separation of the workflow execution and the processing of individual tasks, such as computational jobs or file transfers. The light-weight design is achieved by focusing on the generic workflow execution components and by leveraging state-of-theart Grid technology, e.g., for state management. The current prototype is implemented using the Globus Toolkit 4 (GT4) Java WS Core and has support for executing workflows produced by Karajan. It also includes plugins for task execution with GT4 as well as a high-level Grid job management framework. 1

Abstract. We present an approach for development of Grid resource management tools, where we put into practice internationally established high-level views of future Grid architectures. The approach addresses fundamental Grid challenges and strives towards a future vision of the Grid where capabilities are made available as independent and dynamically assembled utilities, enabling run-time changes in the structure, behavior, and location of software. The presentation is made in terms of design heuristics, design patterns, and quality attributes, and is centered around the key concepts of co-existence, composability, adoptability, adaptability, changeability, and interoperability. The practical realization of the approach is illustrated by five case studies (recently developed Grid tools) high-lighting the most distinct aspects of these key concepts for each tool. The approach contributes to a healthy Grid ecosystem that promotes a natural selection of “surviving ” components through competition, innovation, evolution, and diversity. In conclusion, this environment facilitates the use and composition of components on a per-component basis. 1

Grid computing has emerged as a global cyber-infrastructure for the next-generation of e-Science applications by integrating large-scale, distributed and heterogeneous resources. Scientific communities are utilizing Grids to share, manage and process large data sets. In order to support complex scientific experiments, distributed resources such as computational devices, data, applications, and scientific instruments need to be orchestrated while managing the application workflow operations within Grid environments. This thesis investigates properties of Grid workflow management systems, presents a workflow engine and algorithms for mapping scientific workflow applications to Grid resources based on specified QoS (Quality of Service) constraints. To address the field of Grid computing of workflow application scheduling, the thesis has made the following contributions: • proposed a taxonomy of workflow management systems for Grid computing. • developed a workflow engine which leverages tuple spaces to provide event-based execution management. • developed deadline and budget distribution strategies based on the workload and dependency of tasks. • developed algorithms for scheduling workflows with QoS constraints using genetic algorithms. • leveraged multi-objective evolutionary algorithms (MOEAs) for workflow execution planning to generate a set of trade-off alternative scheduling solutions.

In e-Science environments, scientific workflow management systems (SWMS) hide the integration details among Grid resources and allow scientists to prototype an experimental computing system at a high level of abstraction. However, the development of an effective SWMS requires profound knowledge on both application domains and the network programming, and is often time consuming. Agent technologies provide suitable solutions to decompose the control intelligence of flow execution and to encapsulate distributed e-Science resources. The work presented in this paper is conducted in the context of the Dutch Virtual Laboratory for e-Science (VL-e) project. Agent technologies are proposed to realise generic workflow support. 1