NSS NSB Brief - PowerPoint PPT Presentation

1 / 8
About This Presentation
Title:

NSS NSB Brief

Description:

Only optimistic parallel-processing engine receiving significant DoD funding ... For AFRL, Phase II SBIR will lead to dramatic new simulation capability ... – PowerPoint PPT presentation

Number of Views:28
Avg rating:3.0/5.0
Slides: 9
Provided by: billst5
Category:
Tags: nsb | nss | brief | funding | sbir

less

Transcript and Presenter's Notes

Title: NSS NSB Brief


1
Simulation Sciences Division High Performance
Computing
SPEEDESSynchronous Parallel Environment for
Emulation and Discrete-Event Simulation
2
Overview of SPEEDES
  • METRON Simulation Sciences product since 1996
  • Powerful optimistic-processing parallel
    processing engine
  • Simulation events are distributed over multiple
    processors
  • Events assigned to a processor proceed
    optimistically (i.e. assuming that processing on
    other nodes wont invalidate)
  • When messages from one node invalidate processing
    on other nodes, incorrect processing is rolled
    back, invalid messages recalled, and are
    corrected
  • Result as long as enough processors are
    available, speed and complexity of simulation is
    not limited
  • Processing overhead (typically 15) associated
    with rollback/callback/reprocessing is generally
    not affected by number of processors
  • Parallel-processing simulations mimic real-life,
    where most events happen independently in parallel

3
SPEEDES Development and Use
  • Developed, maintained, and distributed by METRON
  • Development funded by Joint National Integration
    Center (JNIC) and Air Force Research Laboratory
    (AFRL)
  • Metron supplies source code to qualified (U.S.
    only) users at no charge
  • On-line documentation and change request system
  • Primary users
  • JNIC Missile Defense Wargaming and Analysis
    Resource (MDWAR)
  • Air Force Research Lab Joint Battlespace
    Infosphere Simulation (JBISim)
  • Current Development Projects
  • JNIC Multiple-simulation clusters (to add
    THAADS, Patriot, Airborne Laser, Aegis, etc.)
  • AFRL Multiple parallel course-of-action
    simulation capability (Phase II SBIR)

4
Key SPEEDES Developers
Ron Van Iwaarden PhD, Applied Mathematics,
University of Colorado MS, Applied Mathematics,
University of Colorado BS, Mathematics,
University of Colorado SPEEDES developer for 8
years Lead, SPEEDES development and
documentation MDWAR wargame support at the
JNIC
Gary Blank MS, Comp. Science, University of
Virginia BS, Applied Mathematics, Brown
University SPEEDES developer for 8 years Lead,
SPEEDES Multi-COA Enhancement project SPEEDES
enhancements for MDWAR SPEEDES support of AFRL
(DIEMS, GIEsim) SPEEDES FAAsim prototype HLA
RTI developer HLA Federations
Jacob Burckhardt BS, Computer Science, UC
Berkeley SPEEDES developer for 8 years Lead,
JSIMS sim engine IVV SPEEDES enhancements for
MDWAR SPEEDES testing SPEEDES Configuration
Management
5
SPEEDES Funding
(thousands of dollars)
1.0
2.0
2.1
2.2
Version
6
Competition
  • SPEEDES maturity and performance puts it at the
    head of the pack
  • Only optimistic parallel-processing engine
    receiving significant DoD funding
  • RAM Labs has WARP IV derivative, with no known
    users
  • Two MDA Phase II SBIRs starting
  • Georgia Tech Time Warp
  • Used in simulation of communication networks

7
The Future
  • SPEEDES growth at the JNIC
  • The Missile Defense program was severely
    chastised by the for its lack of quality
    simulations by Defense Science Board
  • MDWAR was held up as the sole example of success
  • The future for MDA is Ballistic Missile Defense
    System (BMDS) sim
  • Decision isnt final, but it appears that MDWAR
    will be enhanced to become the BMDS sim
  • Expectation is that METRON will receive at least
    a moderate amount of additional development work
  • For AFRL, Phase II SBIR will lead to dramatic new
    simulation capability
  • Northrop Grumman and Lockheed Martin are
    candidates for Phase III sponsorship

8
Conclusion
  • SPEEDES is excellent choice for all parallel
    simulations
  • Development on SPEEDES continues although at a
    slower pace
  • Future is bright for BMDS simulation
  • Should bring additional funding such as
    clustering or HLA capability
  • Multiple course-of-action capability should make
    SPEEDES of interest to more clients
  • The JNIC and AFRL continue to provide stable and
    consistent funding
Write a Comment
User Comments (0)
About PowerShow.com