skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: CUDA GPU Programming Applied to HSI Exploitation


This presentation summarizes the architecture and heirarchy of GPU/CPA programming.

  1. Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Publication Date:
Research Org.:
Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
Report Number(s):
DOE Contract Number:
Resource Type:
Technical Report
Country of Publication:
United States
97 MATHEMATICS AND COMPUTING; CUDA GPU hyperspectral "remote sensing"

Citation Formats

Armstrong, Derek Elswick. CUDA GPU Programming Applied to HSI Exploitation. United States: N. p., 2017. Web. doi:10.2172/1341833.
Armstrong, Derek Elswick. CUDA GPU Programming Applied to HSI Exploitation. United States. doi:10.2172/1341833.
Armstrong, Derek Elswick. Thu . "CUDA GPU Programming Applied to HSI Exploitation". United States. doi:10.2172/1341833.
title = {CUDA GPU Programming Applied to HSI Exploitation},
author = {Armstrong, Derek Elswick},
abstractNote = {This presentation summarizes the architecture and heirarchy of GPU/CPA programming.},
doi = {10.2172/1341833},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Thu Jan 26 00:00:00 EST 2017},
month = {Thu Jan 26 00:00:00 EST 2017}

Technical Report:

Save / Share:
  • Two spectrally calibrated tarpaulins (tarps) were co-located at a fixed Global Positioning System (GPS) position on the gravel antenna field at the U.S. Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) Climate Research Facility’s Southern Great Plains (SGP) site. Their placement was timed to coincide with the overflight of a new hyperspectral imaging satellite. The intention was to provide an analysis of the data obtained, including the measured and retrieved spectral albedos for the calibration tarps. Subsequently, a full suite of retrieved values of H2O column, and the aerosol overburden, were to be compared to those determined by alternate SGPmore » ground truth assets. To the extent possible, the down-looking cloud images would be assessed against the all-sky images. Because cloud contamination above a certain level precludes the inversion processing of the satellite data, coupled with infrequent targeting opportunities, clear-sky conditions were imposed. The SGP site was chosen not only as a target of opportunity for satellite validation, but as perhaps the best coincident field measurement site, as established by DOE’s ARM Facility. The satellite team had every expectation of using the information obtained from the SGP to improve the inversion products for all subsequent satellite images, including the cloud and radiative models and parameterizations and, thereby, the performance assessment for subsequent and historic image collections. Coordinating with the SGP onsite team, four visits, all in 2009, to the Central Facility occurred: • June 6-8 (successful exploratory visit to plan tarp placements, etc.) • July 18-24 (canceled because of forecast for heavy clouds) • Sep 9-12 (ground tarps placed, onset of clouds) • Nov 7-9 (visit ultimately canceled because of weather predictions). As noted, in each instance, any significant overcast prediction precluded image collection from the satellite. Given the long task-scheduling procedures (which were in place for each time period), coupled with other priorities and the satellite lifetime, no alternate SGP images could be obtained.« less
  • We review some of the classic numerical techniques used to analyze contact discontinuities and compare their effectiveness. Several finite difference methods (the Lax-Wendroff method, a Multidimensional Positive Definite Advection Transport Algorithm (MPDATA) method and a Monotone Upstream Scheme for Conservation Laws (MUSCL) scheme with an Artificial Compression Method (ACM)) as well as the finite element Streamlined Upwind Petrov-Galerkin (SUPG) method were considered. These methods were applied to solve the 2D advection equation. Based on our results we concluded that the MUSCL scheme produces the sharpest interfaces but can inappropriately steepen the solution. The SUPG method seems to represent a goodmore » balance between stability and interface sharpness without any inappropriate steepening. However, for solutions with discontinuities, the MUSCL scheme is superior. In addition, a preliminary implementation in a GPU program is discussed.« less
  • At a high level, my research interests center around designing, programming, and evaluating computer systems that use new approaches to solve interesting problems. The rapid change of technology allows a variety of different architectural approaches to computationally difficult problems, and a constantly shifting set of constraints and trends makes the solutions to these problems both challenging and interesting. One of the most important recent trends in computing has been a move to commodity parallel architectures. This sea change is motivated by the industry’s inability to continue to profitably increase performance on a single processor and instead to move to multiplemore » parallel processors. In the period of review, my most significant work has been leading a research group looking at the use of the graphics processing unit (GPU) as a general-purpose processor. GPUs can potentially deliver superior performance on a broad range of problems than their CPU counterparts, but effectively mapping complex applications to a parallel programming model with an emerging programming environment is a significant and important research problem.« less
  • The numerical solution of linear differential equations by linear programming is described. (D.L.C.)