Workers Section

Milestone: Demonstration at TTF (April 3-6, 2002) and Sherwood (April 22-24, 2002) Fusion Science Meetings:
Advanced 3D Visualization of NIMROD data.
Remote Visualization Capability.

Milestone: Demonstration at SuperComputing 2002 (November 17-22, 2002) and APS/DPP (November 11-15, 2002)
What is our focus here?

• Visualization Application
  • Create an openGL based application for 3D graphics
  • Data read from MDSplus for general applicability
  • Start with NIMROD data to focus our effort but our implementation must be general so that this tool is applicable beyond NIMROD
  • Flexible data interface/calculation interface for extension by scientists
  • Evaluate the success of this strategy including tool flexibility and large data sizes in April 2002
• Remote Collabortion Capability
  • Workstation to Workstation, Workstation to Tiled Wall, and Tiled Wall to Tiled Wall all desired
  • What of our work can we demonstrate in April 2002?
  • What capability can we give scientists at this time?
  • Could even demonstrate commercial solutions if it goes dramatically beyond present capabilities

• Visualization Application
  • On large display, NIMROD data retrieved from DIII-D data server - utilize secure transmission developed in Security/Remote work
  • Hardware is a Linux workstation with fast graphics card
  • From either DIII-D or C-Mod MDSplus data server, TRANSP data - demonstrate vis tool flexibility
  • Initial GS2 data - demonstrate vis tool flexibility
• Remote Collabortion Capability
  • Workstation to Workstation, Workstation to Tiled Wall, and Tiled Wall to Tiled Wall all desired
  • What of our work can we demonstrate in April 2002
  • Could even demonstrate commercial solutions if it goes dramatically beyond present capabilities

1. MDSplus data access for all laboratories [All - Dec 14]
   1. Client installed and tested on sample DIII-D data
2. Understanding present state of experimentalists fusion-scientist visualization [Utah, ANL - Jan 15]
   1. ReviewPlus & IDL installed at CSET laboratories [Utah, ANL - Dec 14]
   2. ReviewPlus used to understand present capabilities. CSET people to make recommendations for how the experimentalists' visualization tools can be enhanced. [Utah, ANL - Jan 11]
   3. Take best ideas and develop a plan for quick prototype for feedback. [Utah, ANL, GA - Jan 18]
3. NIMROD data available for initial visualization work [Nov 30]
   1. Large time-dependent NIMROD run loaded into MDSplus [SAIC - Nov 30]
   2. C routine written for sample MDSplus NIMROD data access [GA - Nov 30]
4. Specify computer hardware requirement for fusion scientist [Utah, ANL - Dec 7]
   1. To run visualiation applications - Post Hardware on web and advertise
   2. Are there implications here for remote collaboration work?
   3. Work with fusion scientists to make purchase - discount strategies?
   4. This hardware will not allow temporal evolution and slicing at the same time. Utah visit will give scientists the chance to see this in action on fast hardware.
5. Prototype visualization techniques using NIMROD data as an example [Utah - Jan 14]
   1. Coordinated with NIMROD team with feedback to Utah by Feb 1
      1. When does NIMROD group visit Utah?
6. Create first draft of general visualization application [Utah - Apr 1]
   1. Demonstration at Sherwood and TTF
7. Initial evaluation of new general visualization application [All - May 1]
   1. Create plan to adjust application related to initial feedback from Sherwood and TTF
   2. Evaluate tool with other datasets - GS2 and TRANSP
   3. Evaluate tool with very large dataset
      1. Push the limits to evaluate tool, MDSplus, and network speeds
      2. Identify early any issues associated with large datasets
8. Coordinate visualization application with NIMROD physics work to be presented at Sherwood - Help the NIMROD team to achieve their goals [GA - Jan 25]
   1. DIII-D Experiments
      1. Dylan Brennan and Rob LaHaye scheduled for January 2002
   2. Theory and experiment comparison
      1. Dalton Schnack - 80,000's series DIII-D pulses compared to NIMROD
   3. Alternatives
      1. Carl Sovinec - MST experimental and NIMROD comparison
      2. Carl Sovinec - SSPX experimental and NIMROD comparison
   4. Electronic presentations by NIMROD scientists at Sherwood
      1. Movies of our visualizations - so much more dramatic than stills
9. Initial testing with documentation to facilitate remote collaboration plan [Dec 21]
   1. Chromium
      1. Try to get this working on multiple Princeton display walls with microwave link[PCS, PPPL]
      2. Try to get this working on multiple ANL display walls [ANL]
      3. Collaborative issue with Chromium - How to get mouse/kbd from remote site back to control app?
   2. VNC for workstation to workstation collaboration
      1. Investigate Java based VNC setup program by NCSA [ANL]
      2. Evaluate viability of VNCWALL and scivis based collaboration including workstation to wall[PCS, PPPL]
   3. Use Access Grid for collaboration
      1. should have these at least installed/demonstrable at our sites
      2. ag on tiled display - maybe use vtk on ag?
   4. Toolkits
      1. investigate fltk or other toolkits for cross-platform gui stuff
   5. Techniques for simple 2D plot collaboration
      1. Java based code scivis to read MDSplus data and visualize on tile display wall[PPPL]
      2. Evaluate a modified RPLOT to use scivis graphics for collaborative visualization[PPPL]
      3. IDl - issue of using OpenGL with Chromium due to library issue [ANL]
   6. MPEG [ANL, PCS]
      1. Experiment with multiple simultaneous MPEG streams
      2. bandwidth issue?
   7. wiregl [PPPL, PCS]
      1. Investigate duplication of displays with wiregl on both display walls
      2. Possible use PCS 24 display nodes and PPPL 12 display nodes simultaneously
10. Xplit [ANL]
    1. Enumeration of performance evaluation metrics
       1. Test to understand the relative performance of pixels versus X-protocal in the transmission of pieces
       2. Latency in the environment versus human interaction need to be understood
    2. Develop prototype Xplit architecture based on these metrics
       1. Demonstrate the moving of one static and one streaming application in a seamless desktop environment on a tiled display wall
       2. Static application will be the IDL based tool ReviewPlus
       3. A vic window will be used for the streaming environment
    3. Prototype will lead into the architecture of Xplit for full development in year 2
11. Access Grid [ANL]
    1. Develop ability of an application to be moved from a private desktop space into a shared space
       1. Protytpe system will be developed focusing on only the X-windows environment
       2. Prototype will allow for the transfer of ReviewPlus from a users desktop into a shared space
    2. Modes of interaction will be investigated, as well as implementation ideas in order to reach a final architecture
12. Investigate AG nodes at Fusion labs
    1. MIT [MIT - Jan 15]
       1. Understand AG approach versus H.323 and how data display tools are integrating into AG environment
    2. PPPL [PPPL - Nov 30]
       1. Possible leverage off of existing tiled display wall.
       2. Has this decision been made yet?
    3. GA [GA - Jan 15]
       1. Investigated leveraging new tiled display wall
13. Questions and issues to be folded into all of above
    1. X-windows on tiled display wall promised for year 2 - Xplit
       1. Start, middle, or end of year 2?
       2. Can a part of this work be demonstrated to show progress?
       3. Can any of this be used by scientists before fully completed?
    2. What solutions for workstation to workstation remote collaboration?
       1. Desired by NIMROD team to perform remote one-on-one discussion regarding physics and code debugging
       2. Desired to get visualizations and information back into control rooms; one-on-one
       3. Viable for demonstration in April 2002?
    3. What solutions for workstation to Tiled Display remote collaboration?
       1. Desired by NIMROD for remote team meetings where a number of single individuals connect to one wall in a room with 5 to 10 scientists
       2. Desired to get visualizations and information back into control rooms in a large shared environment
       3. Viable for demonstration in April 2002?
    4. What solutions for Tiled Display to Tiled Display remote collaboration?
       1. Today, from a customer standpoint, this has few interested parties
       2. Demonstrate on existing walls, motivate fusion scientists about its value
       3. But, does this work drive the two tasks above such that work on this task gives us solutions to the other two along the way?
       4. Test on two Princeton walls connected together?
    5. Access Grid
       1. Prototype interface to AG on tiled display
       2. Prototype desktop visualization application with AG
       3. Prototype of AG workspace docking
       4. What does this mean from a practical standpoint? What can the scientists use in year 1?

Implementation Philosophy Focused on November 2002

team contact info | about the fusion grid | fusiongrid research Last modified 01/14/02. Comments? webmaster