New Paradigms in Problem Solving Environments for Scientific - - PowerPoint PPT Presentation

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

1

New Paradigms in Problem Solving Environments for Scientific Computing

George Chin Jr.

• L. Ruby Leung

Karen Schuchardt Deborah Gracio

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

2

Background

I Pacific Northwest National Laboratory is one of nine U.S.

Department of Energy R&D laboratories

I Multi-purpose, multi-program laboratory I Wide range of disciplines from molecular biology to

atmospheric sciences to manufacturing engineering

I Basic and applied research I Information technology focus on supporting computational

needs of scientists

I Specific computational science R&D focus on scientific

problem-solving environments

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

3

What is a Problem-Solving Environment?

A PSE is a computer system that provides all the computational facilities necessary to solve a target class of problems… Overall, they create a framework that is all things to all people: they solve simple or complex problems, support rapid prototyping or detailed analysis, and can be used in introductory education or at the frontiers of science.

• Gallopoulos, Houstic, and Rice, 1998

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

4

Extensible Computational Chemistry Environment

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

5

Biology Workbench

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

6

Distribution of Scientific Computing

PNNL USGS NCAR SDSC UW ANL ORNL LBNL UNC Scripps UC - Boulder

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

7

Access to Computational Resources

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

8

Context of Scientific Problem Solving

Problem Solving involves “higher-order skills that encompass specific processes and abilities”

• Lauren Resnick, 1987

I Scientific processes

• Experiments

I Scientific abilities

• Appropriate level of abstraction
• Knowledge expression and dissemination

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

9

PSE Research and Development

I Laboratory PSE research and development effort I Participatory analysis I Participatory design I Evolve scientific PSE architectures, environments,

and tools

I Target real scientific fields

• Climate change modeling
• Biomolecular research

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

10

Experimental Design and Execution

I Provide high-level visual representation I Provide user control of experiment execution I Support scientists in design of experiments

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

11

Scientist’s Conceptual View of Research Processes

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

12

Scientific Workflow

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

13

High-Level Access to Computational Resources

I High-level access to diverse computational resources I Conform to scientist’s level of understanding and

abstraction

I Place in context of the overriding computational

experiment or model

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

14

Views of Data

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

15

Views of Processes

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

16

Knowledge Management and Dissemination

I Gather procedural and domain knowledge I Facilitate accumulation and sharing of knowledge I Support both synchronous and asynchronous forms of

knowledge gathering dissemination

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

17

Capture of Knowledge

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

18

In-Context Real-Time Collaboration

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

19

Areas of Research and Development

I Visual programming models I Scientific workflow management I User-level data management I Scientific collaboratories I Adaptive environments

• In-context collaboration capabilities
• Automatic resource integration
• Automatic workflow generation

U.S. Department of Energy Pacific Northwest National Laboratory

2/20/200 2

20

Summary – Our Aims

I Providing context for scientific problem solving I Contextual view of experiment I Appropriate level of abstraction for scientists I Automation of intricate, manual computer operations I Accumulation of domain context and knowledge I Expanding base of scientific users