Success Stories

Radar System Design

Results: 50-second FFTs on 200GB matrix with 13 Billion Elements

The Challenge

"Star-P has dramatically accelerated the development of microwave image reconstruction algorithms by allowing us to stay within the MATLAB® programming environment and exploit supercomputing resources for algorithm development and processing throughput. It's an enabler – without it we would not be using parallel hardware, and the increased capabilities fundamentally transform what we are able to do in the Advanced Radar Waveforms & Processing Branch."

Electronics Engineer
Air Force Research Labs

For years the military relied on radar information from land-based facilities and reconnaissance aircraft, but has recently expanded their missions to include satellite-based radar systems.

But where traditional radar images might measure 10 megabytes, satellite radar can easily inundate a facility with terabytes of data daily, dramatically complicating and delaying the analysis process.

Because researchers are skilled at evaluating radar analysis algorithms using MATLAB®, they wanted to preserve the familiarity and interactivity of their desktop environment while taking advantage of the computational power of parallel HPCs. Due to the time-sensitivity of their work, they cannot afford the time required to re-program their algorithms for parallel processing.

Star-P Solution

A key goal of the AFRL is to enable an agile workforce. Lab researchers believe that Star-P's desktop interactivity and ease-of-use will deliver this agility. Star-P enables standard MATLAB® commands and functions to execute in a parallel manner transparently to the user. This preserves the familiar workflow while tackling data sets orders of magnitude larger than they process on their desktops. For AFRL, Star-P combines the critical parallel approaches in one environment: task and data parallelism, backend support and compilation. Star-P integrates a wide range of linear algebra and other routines seamlessly with MATLAB® for the user. The end result gives the user a familiar interface, and ability to use the parallel server's large processor and memory resources.

"Migrating Modeling and Simulation Applications on to High Performance Computers"
Presented at the SPIE Enabling Technologies for Simulation Science X Conference by AFRL's Mark Barnell & Brian Rahn		 "Rapid Algorithm Development for Large-Scale Microwave Imaging using Star-P"
Presented at the SGI User Group by AFRL’s Kevin Magde & Mark Barnell

Summary & Metrics

  • Large and growing data sets
  • FFT-based signal processing algorithms
  • No need to re-write MATLAB® code in C, Fortran, MPI
  • Two-dimensional FFT on large data set (200 GB, matrices with over 13 billion elements) computes in < 50 seconds on a 128-processor SGI Altix server

"Where traditional radar images might measure 10 megabytes, satellite radar can easily inundate a facility with terabytes of data daily, dramatically complicating and delaying the analysis process. Satellite radar can't help our military readiness if we have to wait a long time for the analysis results. With Star-P, researchers can now preserve their familiar workflow while tackling data sets an order of magnitude larger than they process on their desktops."

-Mark Barnell
Air Force Research Labs