GPU-oriented fine grit parallel application mapping method
An application mapping, fine-grained technology, applied in the direction of concurrent instruction execution, machine execution device, general-purpose stored program computer, etc., can solve problems such as inability to fully utilize application features, low efficiency of heterogeneous systems, etc., and achieve simple and efficient task division. Avoid the effects of unclear task division and small parallel granularity
Inactive Publication Date: 2014-09-17
NAT UNIV OF DEFENSE TECH
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0025] The technical problem to be solved by the present invention is to propose a GPU-oriented fine-grained parallel application mapping method to solve the defects that the random determination of the calculation granularity of the segment-level parallel method leads to low efficiency of the heterogeneous system composed of GPU+CPU and the inability to fully utilize the characteristics of the application program
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0073]By adopting the method of the present invention, a fine-grained pixel-level parallel application mapping method for an n-dimensional data structure input set can be realized, and its flow is as follows figure 2 shown. Methods as below:
[0074] The first step is to build a heterogeneous system composed of GPU+CPU. The specific method is to select a GPU graphics card (such as NVidia’s graphics card) that supports the CUDA architecture, and install it in the graphics card PCI-E slot on the computer motherboard. In this way, the construction of the heterogeneous system composed of GPU+CPU is completed. The microarchitecture of the successfully built GPU+CPU heterogeneous system is as follows: figure 1 shown. figure 1 The right half of the figure is a macroscopic schematic diagram of the CPU+GPU heterogeneous system; the left half is a schematic diagram of the internal microarchitecture of the GPU. The data interaction between the CPU and the GPU is completed through the...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a GPU (Graphic Processing Unit)-oriented fine grit parallel application mapping method, and aims at solving the defects of low efficiency and failure in fully using the characteristics of the application programs of a heterogeneous system formed by GPU+CPU (Central Processing Unit), caused by randomly determining the calculation grit in the conventional mapping method. The technical scheme is as follows: the GPU -oriented fine grit parallel application mapping method comprises the following steps of: building the heterogeneous system formed by GPU+CPU, installing the GPU operating environment and starting the GPU to execute initialization, wherein the CPU is in charge of inputting application mapping into an internal memory and starting an API (Application Program Interface) function to input the application mapping to a GPU video memory; for a multi-dimensional array, firstly, converting the multi-dimensional array into a two-dimensional array, pixelating the two-dimensional array, and finally, using the GPU+CPU heterogeneous calculation system to realize the fine grit pixel level parallel application mapping oriented to an n-dimensional data structure input set. With the adoption of the method provided by the invention, the characteristics of the application program can be fully used, tasks are divided clearly, the parallel grit is small, the efficiency of the GPU+CPU heterogeneous calculation system can be maximized, and the performance speed-up ratio is remarkable.
Description
technical field [0001] The invention relates to an application mapping method in the field of graphics processing unit GPU (Graphics Processing Unit) architecture, in particular to a fine-grained parallel application mapping method for GPU general purpose computing GPGPU (General Purpose computing on Graphics Processing Units). Background technique [0002] The so-called architecture-oriented application mapping is a method for efficiently transplanting a specific application to a specific target architecture. Without an architecture-oriented application mapping approach, a specific application may not run efficiently on the target architecture, or even run on the target architecture at all. Therefore, the architecture-oriented application mapping method is the basis for the research and application of high-performance computing architecture. [0003] With the advancement of science and technology, the branches of scientific applications have become increasingly refined, an...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F9/38G06F15/80
Inventor 王志英甘新标沈立黄立波陆洪毅朱琪苏博郑重徐帆魏雄
Owner NAT UNIV OF DEFENSE TECH