Interface design method, device and system between heterogeneous computing platform subsystems

A technology of interface design and computing platform, applied in cloud processing, machine learning algorithms, and big data fields, can solve the problems of reducing cluster expansion computing performance gains, increasing system costs and energy consumption, etc., to increase system costs and energy consumption, and improve Computational Performance Effects

Active Publication Date: 2019-09-24
ZHENGZHOU YUNHAI INFORMATION TECH CO LTD
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the computing performance is improved, the system cost and energy consumption increase rapidly, which greatly reduces the computing performance benefits brought about by cluster expansion.

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Interface design method, device and system between heterogeneous computing platform subsystems
  • Interface design method, device and system between heterogeneous computing platform subsystems
  • Interface design method, device and system between heterogeneous computing platform subsystems

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] see first figure 2 , figure 2 It is a schematic flowchart of an interface design method between heterogeneous computing platform subsystems provided by an embodiment of the present invention. The embodiment of the present invention may include:

[0054] S201: Make a local statement on the FPGA called method in the local file of the Spark platform.

[0055] The compiled language of the Spark platform can be Scala, and the programming language of FPGA can be OpenCL. By using Scala to call the compiled language (OpenCL) special interface of other subsystems that constitute the heterogeneous platform and compile it into the Spark framework, the realization of the Combination of subsystems and Spark platform. Of course, if necessary, Scala may not be selected as the compilation language, and other applicable languages ​​may be selected, which is not limited in this embodiment of the present invention. However, preferably, the selected language should be mature, compatib...

Embodiment 2

[0076] see image 3 , image 3 The schematic flow diagram of another interface design method between heterogeneous computing platform subsystems provided by the embodiment of the present invention, the embodiment of the present invention can form heterogeneous platform with FPGA for example, realize Scala language and OpenCL in JNI (Java NativeInterface) Language interaction, specifically, may include the following:

[0077] S301: The Scala language performs local declaration on the method calling OpenCL.

[0078] S302: Recompile the modified Scala language according to preset instructions.

[0079] S303: Using the javah command to generate a header file containing the method for calling OpenCL.

[0080]S304: Compile the OpenCL code according to the header file to generate an so library file, and copy the library file to java.library.path.

[0081] Wherein, for explanations of S301-S304, reference may be made to the explanations of S201-S204, which will not be repeated her...

Embodiment 3

[0089] see Figure 4 , Figure 4 A structural diagram of an interface design device between heterogeneous computing platform subsystems provided by an embodiment of the present invention, the device may include:

[0090] The declaration module 401 is used for making a local declaration on the FPGA called method in the local file of the Spark platform.

[0091] The compiling module 402 is configured to recompile the files of the modified Spark platform according to preset instructions.

[0092] The header file generating module 403 is configured to use preset commands to generate a header file containing the method called by the FPGA.

[0093] The library file generation module 404 is used to generate a dynamic library file according to the FPGA called method and the header file, and copy the dynamic library file to a preset path to complete the interface design between the Spark platform and the FPGA.

[0094] Optionally, in some implementations of this embodiment, please r...

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

No PUM Login to view more

Abstract

The embodiment of the invention discloses a method for designing an interface between heterogeneous computing platforms. The method comprises the following steps: performing local statement by using a FPGA called method through a local file of a Spark platform, and then compiling the modified Spark file again; generating a header file containing the FPGA called method by using a preset command, generating a dynamic library file in combination with the FPGA called method, and finally copying the file to a preset path, and then the interface between the Spark platform and the FPGA is designed. The effective combination of subsystems constituting the heterogeneous platform is realized, the heterogeneous computing platforms operate healthy, the problem of overlarge cost of the internal memory of the Spark platform is effectively solved, the data processing speed is accelerated, and the computing performance of the Spark platform is improved beneficially. In addition, a corresponding implementation device and system is further provided to further improve the practicability of the method, and the device and the system have corresponding advantages.

Description

technical field [0001] The present invention relates to the fields of big data, cloud processing and machine learning algorithms, in particular to an interface design method, device and system between heterogeneous computing platform subsystems. Background technique [0002] With the advent of the Internet of Things and 5G communication era, the field of big data is also facing tremendous changes. Data processing is developing in the direction of intelligent data mining, which greatly promotes the research and application of machine learning. [0003] ML (Machine Learning, Machine Learning) is a multi-field interdisciplinary subject, specializing in the study of how machines simulate or implement human learning behaviors to acquire new knowledge or skills, and reorganize existing knowledge structures to continuously improve their own performance. . It is the core of artificial intelligence and the fundamental way to make computers intelligent. Its application pervades all f...

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
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06F8/20
Inventor 曹芳陈继承王洪伟
Owner ZHENGZHOU YUNHAI INFORMATION TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap