A Survey of General‐Purpose Computation on Graphics Hardware

  title={A Survey of General‐Purpose Computation on Graphics Hardware},
  author={John Douglas Owens and David P. Luebke and Naga K. Govindaraju and Mark J. Harris and Jens H. Kr{\"u}ger and Aaron E. Lefohn and Timothy J. Purcell},
  journal={Computer Graphics Forum},
The rapid increase in the performance of graphics hardware, coupled with recent improvements in its programmability, have made graphics hardware a compelling platform for computationally demanding tasks in a wide variety of application domains. In this report, we describe, summarize, and analyze the latest research in mapping general‐purpose computation to graphics hardware. 
Recent trends in software and hardware for GPGPU computing: A comprehensive survey
The technical motivations that underlie GPU computing are given and the hardware and software developments that have led to the recent interest in this field are described.
Comparison of Technologies for General-Purpose Computing on Graphics Processing Units
The computational capacity of graphics cards for general-purpose computinghave progressed fast over the last decade. A major reason is computational heavycomputer games, where standard of performan
Attaining High Performance in General-Purpose Computations on Current Graphics Processors
This paper evaluates the performance of linear algebra and image processing routines, both on classical and unified GPU architectures and traditional processors (CPUs).
General-Purpose Computation Using Graphics Hardware for Fast HDR Image Processing
This paper presents a new approach to hardware aided image processing and analysis, primarily focused on HDR imaging, which proposes a library model which is able to utilize efficiently most powerful of the underlying vector hardware.
An Introduction to General-Purpose Computing on Programmable Graphics Hardware
The architecture and programming model of programmable graphics hardware is explained, and examples of GPGPU-research at SINTEF within simulation, visualization, image processing, and geometry processing are presented.
Rise of the Graphics Processor
  • D. Blythe
  • Computer Science
    Proceedings of the IEEE
  • 2008
This work examines some of this evolution of hardware to accelerate graphics processing operations, looks at the structure of a modern GPU, and discusses how graphics processing exploits this structure and how nongraphical applications can take advantage of this capability.
General Purpose Programming on Modern Graphics Hardware
The motivations for GPGPU programming, and the capabilities of modern GPUs (including advantages and disadvantages), are explored, as well as the terminology used and the resources available.
GPU-CPU multi-core for real-time signal processing
  • S. Mohanty
  • Computer Science
    2009 Digest of Technical Papers International Conference on Consumer Electronics
  • 2009
This paper analyzes the computation of general-purpose algorithms on GPU, which is a tremendous feat of highly parallel SIMD processing, but its capability is often under-utilized.
Architecting graphics processors for non-graphics compute acceleration
  • Tor M. Aamodt
  • Computer Science
    2009 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing
  • 2009
This paper discusses the emergence of graphics processing units (GPUs) that contain architecture features for accelerating non-graphics (or GPGPU) applications. It provides an introduction for those
Computation on Programmable Graphics Hardware
This article describes approaches to using GPU processing power to accelerate traditionally CPU-based tasks and discusses a specific GPU image-processing application that is a common postprocess for many physically based rendering systems.


Stack Implementation on Programmable Graphics Hardware
A technique is presented that allows the implementation of a stack on programmable graphics hardware, using textures and fragment shaders, which enables a whole new class of GPU algorithms, including recursive functions on complex data structures.
Fast Image Segmentation and Smoothing Using Commodity Graphics Hardware
Preliminary results show a performance increase of over 30% using an nVidia GeForce4 when compared to an implementation using Intel MMX optimized code on a 2.2 Ghz Intel P4 CPU.
Graphics Processor Units: New Prospects for Parallel Computing
This chapter provides an introduction to the use of Graphics Processor Units (GPUs) as parallel computing devices. It describes the architecture, the available functionality and the programming
Fast Matrix Multiplies Using Graphics Hardware
We present a technique for large matrix-matrix multiplies using low cost graphics hardware. The result is computed by literally visualizing the computations of a simple parallel processing algorithm.
General Calculations using Graphics Hardware with Applications to Interactive Caustics
It is shown that graphics hardware can perform general calculations, which accelerate the rendering process much earlier than at the latter image generation stages, an example of the real time calculation of refractive caustics.
Parallel computers for graphics applications
Two processors are presented that use parallel SIMD data paths to support common graphics data structures as primitive operands in arithmetic expressions to provide better price/performance than general purpose designs.
Using modern graphics architectures for general-purpose computing: a framework and analysis
This paper develops a programming framework and applies it to a variety of problems, including matrix multiplication and 3-SAT, and believes that efficient use of graphics hardware will become increasingly important to high-performance computing on commodity hardware.
The GeForce 6800
The architectural drivers for GPUs are programmability, parallelism, bandwidth, and memory characteristics, and this article describes how one team approached the design problem.
GPGPU: general-purpose computation on graphics hardware
The graphics processor (GPU) on today's commodity video cards has evolved into an extremely powerful and flexible processor. Modern graphics architectures provide tremendous memory bandwidth and
Kohonen Feature Mapping through Graphics Hardware
This work describes the utilization of the inherent parallelism of commonly available hardware graphics accelerators for the realization of the Kohonen feature map. The result is an essential