In this paper we address the problem of automatically generating job workflows for the Grid. These workflows describe the execution of a complex application built from individual application components. In our work we have developed two workflow generators: the first (the Concrete Workflow Generator CWG) maps an abstract workflow defined in terms of application-level components to the set of available Grid resources. The second generator (Abstract and Concrete Workflow Generator, ACWG) takes a wider perspective and not only performs the abstract to concrete mapping but also enables the construction of the abstract workflow based on the available components. This system operates in the application domain and chooses application components based on the application metadata attributes. We describe our current ACWG based on AI planning technologies and outline how these technologies can play a crucial role in developing complex application workflows in Grid environments. Although our work is preliminary, CWG has already been used to map high energy physics applications onto the Grid. In one particular experiment, a set of production runs lasted 7 days and resulted in the generation of 167,500 events by 678 jobs. Additionally, ACWG was used to map gravitational physics workflows, with hundreds of nodes onto the available resources, resulting in 975 tasks, 1365 data transfers and 975 output files produced.

Abstract not found

Grid computing provides key infrastructure for distributed problem solving in dynamic virtual organizations. It has been adopted by many scientific projects, and industrial interest is rising rapidly. However, Grids are still the domain of a few highly trained programmers with expertise in networking, high-performance computing, and operating systems. This paper describes our initial work in capturing knowledge and heuristics about how to select application components and computing resources, and using that knowledge to generate automatically executable job workflows for the Grid. Our system is implemented and integrated with a Grid environment where it has generated dozens of workflows with hundreds of jobs in real time. The paper

Planning and scheduling have traditionally been considered separately. Starting from an ordered set of actions or plan, the project manager arranges the actions in time so that temporal and resource constraints are satisfied. However the benefits resulting from the integration of both processes merit an extensive work in this direction. In this paper an integration model for planning and scheduling is proposed. The application of that model to a rescue problem points out the flexibility added to the scheduling process.

I reflect on the progress that the AI Planning field has made over the past 30 years and define the directions where I believe that we should focus our future efforts if we are to continue as a successful and vibrant scientific field. More concretely, I argue that Fikes and Nilsson's original framing of the planning problem is too far removed from the requirements of real-world applications. We must accept that it will not be feasible to obtain complete and consistent domain theories in the near future and that users will want to influence the plans they receive in dimensions other than goal state and shortest path. In light of this, I argue that we must focus on computer-aided planning instead of computer-replaced planning. I outline the technical challenges this course offers and the exciting work already emerging.

This paper considers the problem of generating conflict-free movement schedules for a set of vehicles that are operating simultaneously in a common airspace. In both civilian air traffic management and military air campaign planning contexts, it is crucial that the movements of different vehicles be coordinated so as to avoid collisions and near misses. Our approach starts from a view of airspace management as a four-dimensional resource allocation problem, where the space in which vehicles must maneuver is itself managed as a capacitated resource. We introduce a linear octree representation of airspace capacity to index vector-based vehicle routes and efficiently detect regions of potential conflict. Generalizing the notion of contention-based search heuristics, we next define a scheduling algorithm that proceeds by first solving a relaxed version of the problem to construct a spatial capacity profile (represented as an octree), and then using spatio-temporal regions where demand exceeds capacity to make conflict-avoiding vehicle routing and scheduling decisions. We illustrate the utility of this basic representation and search algorithm in two ways. First, to demonstrate the overall viability of the approach, we present experimental results using data notion of "encounter set", which tolerates some amount of conflict on the assumption that onboard deconfliction processes can take appropriate avoidance maneuvers at execution time, and show that generation of this more abstract form of predictive guidance can be obtained without loss in computational efficiency.

Grid computing provides key infrastructure for distributed problem solving in dynamic virtual organizations. It has been adopted by many scientific projects, and industrial interest is rising rapidly. However, grids are still the domain of a few highly trained programmers with expertise in networking, high-performance computing, and operating systems. We have been working in capturing knowledge and heuristics about how to select application components and computing resources, and using that knowledge to generate automatically executable job workflows for a grid. Our system is implemented and integrated with a grid environment where it has generated dozens of workflows with hundreds of jobs in real time. In order to be applicable to a wide range of existing and new grid applications, the planner needs to be able to work with varying levels of semantic information for processes and the information they consume and create. We discuss our experiences dealing with different levels of data and describe a planning-based system that can provide different levels of support based in the information available.