Most modern computers or game consoles are equipped with powerful yet cost-effective Graphics Processing Units (GPUs) to accelerate graphics operations. Though the graphics engines in these GPUs are specially designed for graphics operations, can the authors harness their computing power for more general nongraphics operations? The answer is positive. This paper presents the study on leveraging the GPUs graphics engine to accelerate the video decoding. Specifically, a video decoding framework that involves both the Central Processing Unit (CPU) and the GPU is proposed. By moving the whole motion compensation feedback loop of the decoder to the GPU, the CPU and GPU have been made to work in parallel in a pipelining fashion.
Did you find this white paper useful?
26 out of 50 users found this white paper useful
Related white papers
Walsh Finite Scheduling System
The Walsh Finite Scheduling System generates a Dynamic Production Schedule for batch processes used in the pharmaceutical, chemical and bulk pharmaceutical industries. The scheduler takes into account material resources, availability...
Highly Parallel System Architecture in the Compaq Professional Workstation SP700
As business-critical applications become more demanding, the need for more bandwidth becomes increasingly important to customers. The second-generation Highly Parallel System Architecture implemented in the Compaq Professional Workstation SP700 improves...
Reliable Parallel File System Using RAID Technology
Providing data availability in a cluster environment is very important. Most clusters either use RAID technology or redundant nodes to ensure reliability. High performance clusters usually use parallel file systems...
Japan WAM Supports Its Aging Population with Help from IBM
The population of Japan is aging at a rate unprecedented in other nations. It is expected that by 2010, 25 percent of the Japanese population will be over 65. There...
Standard Life: Creating the Infrastructure to Take Advantage of Changing Business Conditions
Standard Life of the U.K. is a 176-year-old company that is vaulting itself into the 21st century by creating an advanced e-business infrastructure that will provide significant benefits to its...
Parallel Transient Non-Linear Analysis of Reinforced Concrete Plates
This paper shows how parallel processing may improve the computational efficiency of the transient dynamic non-linear analysis of reinforced concrete plates subjected to blast loading. The results presented in the...
WIPL-D Parallelization Effort
This paper presents the results of the final year of a development effort to provide a scalable, portable, parallel scene generation tool that provides the capability to rapidly generate scenes...
