Pierre Mouallem

Tracking Files in the Kepler Provenance Framework

Workflow Management Systems (WFMS), such as Kepler, are proving to be an important tool in scientific problem solving. They can automate and manage complex processes and huge amounts of data produced by petascale simulations. Typically, the produced data need to be properly visualized and analyzed by scientists in order to achieve the desired scientific goals. Both run-time and post analysis may benefit from, even require, additional meta-data --- provenance information. One of the challenges in this context is the tracking of the data files that can be produced in very large numbers during stages of the workflow, such as visualizations. The Kepler provenance framework collects all or part of the raw information flowing through the workflow graph. This information then needs to be further parsed to extract meta-data of interest. This can be done through add-on tools and algorithms. We show how to automate tracking specific information such as data files locations.

Toward a first-principles integrated simulation of tokamak edge plasmas

Performance of the ITER is anticipated to be highly sensitive to the edge plasma condition. The edge pedestal in ITER needs to be predicted from an integrated simulation of the necessary first-principles, multi-scale physics codes. The mission of the SciDAC Fusion Simulation Project (FSP) Prototype Center for Plasma Edge Simulation (CPES) is to deliver such a code integration framework by (1) building new kinetic codes XGC0 and XGC1, which can simulate the edge pedestal buildup; (2) using and improving the existing MHD codes ELITE, M3D-OMP, M3D-MPP and NIMROD, for study of large-scale edge instabilities called Edge Localized Modes (ELMs); and (3) integrating the codes into a framework using cutting-edge computer science technology. Collaborative effort among physics, computer science, and applied mathematics within CPES has created the first working version of the End-to-end Framework for Fusion Integrated Simulation (EFFIS), which can be used to study the pedestal-ELM cycles.

Collaboration portal for petascale simulations

The emergence of petascale computing is creating a tsunami of data from peta-scale simulations. Typically, results are analyzed by dozens of scientists who often work as teams. Obviously, it is very important to help these teams by facilitating management, analysis, sharing, and visualization of the data produced by their simulations, and by the auxiliary programs and activities used in the scientific process. One aspect of this is leveraging of their collective knowledge and experiences through a scientific social network. This can be achieved through a combination of back-end IT services, provenance capturing, and easy to use front-end tools. “eSimMon”, is one such tool. In this paper we describe this analysis support system, discuss its ease of use, its efficiency, and its ability to accelerate scientific discovery.

Automation of Network-Based Scientific Workflows

Comprehensive, end-to-end, data and workflow management solutions are needed to handle the increasing complexity of processes and data volumes associated with modern distributed scientific problem solving, such as ultrascale simulations and high-throughput experiments. The key to the solution is an integrated network-based framework that is functional, dependable, faulttolerant, and supports data and process provenance. Such a framework needs to make development and use of application workflows dramatically easier so that scientists’ efforts can shift away from data management and utility software development to scientific research and discovery. An integrated view of these activities is provided by the notion of scientific workflows - a series of structured activities and computations that arise in scientific problem-solving. An information technology framework that supports scientific workflows is the Ptolemy II based environment called Kepler. This paper discusses the issues associated with practical automation of scientific processes and workflows and illustrates this with workflows developed using the Kepler framework and tools.

Fault-Tolerance in Dataflow-Based Scientific Workflow Management

This paper addresses the challenges of providing fault-tolerance in scientific workflow management. The specification and handling of faults in scientific workflows should be defined precisely in order to ensure the consistent execution against the process-specific requirements. We identified a number of typical failure patterns that occur in real-life scientific workflow executions. Following the intuitive recovery strategies that correspond to the identified patterns, we developed the methodologies that integrate recovery fragments into fault-prone scientific workflow models. Compared to the existing fault-tolerance mechanisms, the propositions reduce the effort of workflow designers by defining recovery fragments automatically. Furthermore, the developed framework implements the necessary mechanisms to capture the faults from the different layers of a scientific workflow management architecture. Experience indicates that the framework can be employed effectively to model, capture and tolerate the typical failure patterns that we identified.