Date: Monday, October 4th 10:00 - 12:00 AM
Location: 13-301

Title:Tutorial on Using the NetLogger Toolkit 2.x
Presenter:

Dan Gunter

Lawrence Berkeley National Laboratory

Computer Scientist

Abstract: We will give a tutorial on the major components of the NetLogger Toolkit. The NetLogger Toolkit (short for Networked Application Logger) is a software package that aids in the debugging and performance analysis of complex distributed computer science applications (e.g. Grid computing). It includes tools for application instrumentation, host and network sensors, log publication and collection, and exploratory visual performance analysis. The approach is novel in that it combines network, host, and application-level monitoring to provide a complete view of the entire system. This has proven invaluable for isolating and correcting performance bottlenecks and debugging distributed applications.

The NetLogger Toolkit

Date: Tuesday, October 5th 2:00 - 3:30 PM
Location: 13-215

Title:Recommendations for the Fusion Simulation Project
Presenter:

Douglass E. Post

Los Alamos National Laboratory

Chair, US DOE Fusion Simulation Project Steering Committee

• Seminar Viewgraphs
• Report of the Fusion Simulation Project Steering Committee

Abstract: The proposed Fusion Simulation Project is a multi-institutional computational science project designed to develop a predictive capability for burning plasma experiments with a focus on ITER. The Fusion Simulation Project Steering Committee was chartered by DOE to develop an initial set of requirements and plans for the project based on the initial vision developed by the FESAC ISOFS (Dahlburg committee) report. It would consist of three major elements:

1. A production component that would develop and support suites of codes for use by the tokamak community. It would include an integrated model for tokamak performance.

2. An integration and research component that would further the development of the physics simulation capability in areas such as turbulent transport and MHD needed for an integrated model, and the integration of physics models across multiple time and distance scales.

3. A software and collaboration infrastructure that would be necessary to support the project. The committee recommended that DOE establish a team to conduct a conceptual design for the project, and that if the conceptual design activity is judged successful, that the DOE proceed with the project. The project will begin with about $2M for the first year of conceptual design and ramp up in steps to about $20M per year for 15 years to carry out the whole project. The project would be supported jointly by OFES and OASCR.