Prof. John R. Humphrey Profile

Prof. John R. Humphrey

at EM Photonics Inc

SPIE Involvement:

Author | Instructor

Proceedings Article | 13 June 2014 Paper

Optimization techniques for OpenCL-based linear algebra routines

Stephen Kozacik, Paul Fox, John Humphrey, Aryeh Kuller, Eric Kelmelis, Dennis Prather

Proceedings Volume 9095, 90950D (2014) https://doi.org/10.1117/12.2050673

KEYWORDS: Linear algebra, Field programmable gate arrays, Matrix multiplication, Matrices, Standards development, Algorithm development, Optimization (mathematics), Graphics processing units, Computer programming, Image processing

Read Abstract +

Proceedings Article | 13 June 2014 Paper

Targeting multiple heterogeneous hardware platforms with OpenCL

Paul Fox, Stephen Kozacik, John Humphrey, Aaron Paolini, Aryeh Kuller, Eric Kelmelis

Proceedings Volume 9095, 90950E (2014) https://doi.org/10.1117/12.2050643

KEYWORDS: Control systems, Computer programming, Switching, Parallel computing, Manufacturing, Standards development, Wavefronts, Detection and tracking algorithms, Algorithm development, Photonics

Read Abstract +

The OpenCL API allows for the abstract expression of parallel, heterogeneous computing, but hardware implementations have substantial implementation differences. The abstractions provided by the OpenCL API are often insufficiently high-level to conceal differences in hardware architecture. Additionally, implementations often do not take advantage of potential performance gains from certain features due to hardware limitations and other factors. These factors make it challenging to produce code that is portable in practice, resulting in much OpenCL code being duplicated for each hardware platform being targeted. This duplication of effort offsets the principal advantage of OpenCL: portability. The use of certain coding practices can mitigate this problem, allowing a common code base to be adapted to perform well across a wide range of hardware platforms. To this end, we explore some general practices for producing performant code that are effective across platforms. Additionally, we explore some ways of modularizing code to enable optional optimizations that take advantage of hardware-specific characteristics. The minimum requirement for portability implies avoiding the use of OpenCL features that are optional, not widely implemented, poorly implemented, or missing in major implementations. Exposing multiple levels of parallelism allows hardware to take advantage of the types of parallelism it supports, from the task level down to explicit vector operations. Static optimizations and branch elimination in device code help the platform compiler to effectively optimize programs. Modularization of some code is important to allow operations to be chosen for performance on target hardware. Optional subroutines exploiting explicit memory locality allow for different memory hierarchies to be exploited for maximum performance. The C preprocessor and JIT compilation using the OpenCL runtime can be used to enable some of these techniques, as well as to factor in hardware-specific optimizations as necessary.

Proceedings Article | 9 June 2014 Paper

Multi-frame image processing with panning cameras and moving subjects

Aaron Paolini, John Humphrey, Petersen Curt, Eric Kelmelis

Proceedings Volume 9076, 907607 (2014) https://doi.org/10.1117/12.2050537

KEYWORDS: Image processing, Cameras, Turbulence, Algorithm development, Video, Video surveillance, Video processing, Surveillance, Image enhancement, Speckle imaging

Read Abstract +

Proceedings Article | 29 May 2013 Paper

Advances in computational fluid dynamics solvers for modern computing environments

Daniel Hertenstein, John Humphrey, Aaron Paolini, Eric Kelmelis

Proceedings Volume 8752, 87520B (2013) https://doi.org/10.1117/12.2018918

KEYWORDS: Chemical elements, Computational fluid dynamics, Data communications, Computer architecture, Data processing, Photonics, Profiling, Defense and security, Computer simulations, Parallel computing

Read Abstract +

Proceedings Article | 21 May 2011 Paper

Accelerating sparse linear algebra using graphics processing units

Kyle Spagnoli, John Humphrey, Daniel Price, Eric Kelmelis

Proceedings Volume 8060, 806004 (2011) https://doi.org/10.1117/12.884169

KEYWORDS: Linear algebra, MATLAB, Matrices, Graphics processing units, Computing systems, Parallel computing, Performance modeling, Mathematics, Chemical elements, Excel

Read Abstract +

Proceedings Article | 1 May 2010 Paper

Analyzing the impact of data movement on GPU computations

Daniel Price, John Humphrey, Kyle Spagnoli, Aaron Paolini

Proceedings Volume 7705, 77050D (2010) https://doi.org/10.1117/12.852632

KEYWORDS: Linear algebra, Computer programming, Motion analysis, Visualization, Photonics, Ecosystems, Standards development, Algorithm development, Parallel computing, Computer programming languages

Read Abstract +

Proceedings Article | 27 April 2010 Paper

CULA: hybrid GPU accelerated linear algebra routines

John Humphrey, Daniel Price, Kyle Spagnoli, Aaron Paolini, Eric Kelmelis

Proceedings Volume 7705, 770502 (2010) https://doi.org/10.1117/12.850538

KEYWORDS: Linear algebra, Graphics processing units, Computing systems, Parallel computing, Mathematics, Performance modeling, Computer programming, Matrices, Statistical analysis, Photonics

Read Abstract +

Proceedings Article | 30 April 2009 Paper

A GPU-accelerated toolbox for the solutions of systems of linear equations

John Humphrey, Aaron Paolini, Daniel Price, Eric Kelmelis

Proceedings Volume 7348, 73480Q (2009) https://doi.org/10.1117/12.818598

KEYWORDS: Computing systems, Chemical elements, Graphics processing units, Field programmable gate arrays, Computer programming, Matrices, Photonics, Visualization, Standards development, Computer architecture

Read Abstract +

Proceedings Article | 11 April 2008 Paper

Accelerated determination of UAV flight envelopes

Michael Bodnar, Lyle Long, John Humphrey, Eric Kelmelis

Proceedings Volume 6965, 69650F (2008) https://doi.org/10.1117/12.777309

KEYWORDS: Unmanned aerial vehicles, Computer simulations, Modeling and simulation, Visualization, System integration, Computational fluid dynamics, Graphics processing units, Quantum wells, CFD analysis, Mathematical modeling

Read Abstract +

Proceedings Article | 22 March 2007 Paper

GPU-based accelerated 2D and 3D FDTD solvers

Daniel Price, John Humphrey, Eric Kelmelis

Proceedings Volume 6468, 646806 (2007) https://doi.org/10.1117/12.715044

KEYWORDS: Finite-difference time-domain method, Sensors, Visualization, Electromagnetism, Resonators, Scattering, Computational electromagnetics, Computer simulations, Optical spheres, Radio propagation

Read Abstract +

Proceedings Article | 22 September 2006 Paper

Integrated optical sensor using a dispersion guided photonic crystal structure

Richard Martin, Ahmed Sharkawy, John Humphrey, Eric Kelmelis, Dennis Prather

Proceedings Volume 6322, 63220I (2006) https://doi.org/10.1117/12.681136

KEYWORDS: Sensors, Photonic crystals, Beam splitters, Dispersion, Refractive index, Interferometers, Integrated optics, Chemical reactions, Environmental sensing, Chemical analysis

Read Abstract +

Proceedings Article | 8 September 2006 Paper

Modeling and simulation of nanoscale devices with a desktop supercomputer

Eric Kelmelis, James Durbano, John Humphrey, Fernando Ortiz, Petersen Curt

Proceedings Volume 6328, 632804 (2006) https://doi.org/10.1117/12.681085

KEYWORDS: Field programmable gate arrays, Finite-difference time-domain method, Waveguides, Computer simulations, Visualization, Resonators, Nanophotonics, Computing systems, Modeling and simulation, Photonic crystals

Read Abstract +

Proceedings Article | 22 May 2006 Paper

Accelerated modeling and simulation with a desktop supercomputer

Eric Kelmelis, John Humphrey, James Durbano, Fernando Ortiz

Proceedings Volume 6227, 62270N (2006) https://doi.org/10.1117/12.668281

KEYWORDS: Field programmable gate arrays, Finite-difference time-domain method, Computer simulations, Visualization, Modeling and simulation, Standards development, Electromagnetic radiation, Radio propagation, Computing systems, Data communications

Read Abstract +

Showing 5 of 13 publications

Course Instructor

SC1069: GPU for Defense Applications

This course teaches the basics of utilizing modern programmable graphics processing units (GPUs) for military applications. The modern GPU is a fully programmable parallel programming environment that performs computations an order of magnitude faster than the modern CPU. In this course, we will learn broadly about the architecture of the GPU, the appropriate situations where speedups may be obtained and gain an understanding of the tools and languages that are available for development. Programming is not a part of the curriculum. We will also discuss the available GPU platforms, with an emphasis on rugged, deployable, and low-power offerings. Lastly, the bulk of the course will center on applications and case studies, with emphasis on applications we have produced, including: real-time image processing for the reduction of atmospheric turbulence, applied accelerated linear algebra, image enhancement via super resolution, computational fluid dynamics, and computational electromagnetics.

Expand Description +

Keywords/Phrases

Search In:

Publication Years