Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Heterogeneous accelerating genetic calculating method based on FPGA in cloud platform

A technology for accelerating genes and computing methods, applied in the field of bioinformatics computing, can solve problems such as unsatisfactory application feature diversity, inflexible task allocation, discounted computing efficiency, etc., achieve fast speed, reduce transmission times, and improve acceleration effects Effect

Inactive Publication Date: 2017-05-31
HARBIN INST OF TECH AT WEIHAI
View PDF3 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, simply applying the FPGA heterogeneous accelerated computing system to genetic computing cannot meet the needs of application feature diversity
Purely based on the FPGA heterogeneous acceleration system, the main control processor and FPGA with fixed functions are generally used for calculation, especially the hardware acceleration program in the FPGA, so the task allocation is extremely inflexible, and the software and hardware cannot be automatically performed according to the different genetic calculation programs. Division, which greatly reduces the computational efficiency

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
  • Heterogeneous accelerating genetic calculating method based on FPGA in cloud platform
  • Heterogeneous accelerating genetic calculating method based on FPGA in cloud platform

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] Such as figure 1 Shown, a kind of FPGA heterogeneous acceleration genetic computing method based on the cloud platform of the present invention mainly comprises the following steps:

[0021] Step S1, algorithm analysis, distinguishes the gene calculation into the slow part calculated by the computer processor and the fast part calculated by the computer processor.

[0022] In step S2, according to the algorithm analysis result in step S1, hardware acceleration is used for the slow computing part of the computer processor, and software acceleration is used for the fast computing part of the computer processor.

[0023] Step S3, according to the division of labor in step S2, the software acceleration part is realized through the cloud platform, and the hardware acceleration part is downloaded to the FPGA provided by the cloud platform for realization.

[0024] Step S4, integrating heterogeneous acceleration results.

[0025] In this implementation, the present invention...

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 invention discloses a heterogeneous accelerating genetic calculating method based on FPGA in a cloud platform. By the calculation method, genetic calculation is divided into a slow calculation part in a computer processor and a fast calculation part in the computer processor, hardware is adopted to accelerate according to the slow calculation part, software is adopted to accelerate according to the fast calculation part, the software accelerating part is achieved by the cloud platform, the hardware accelerating part is achieved by being downloaded to the FPGA provided by the cloud platform, and finally the acceleration is achieved by synthesizing results of heterogeneous acceleration. According to the method, heterogeneous calculation used by both a universal calculating platform and the FPGA is adopted, compared with a single processing unit, the FPGA can achieve the optimal combination of a hardware structure and calculation by changing the connecting relations of the hardware, and the speed is faster; compared with calculation which only uses the FPGA, the heterogeneous calculation enables processing units of two structures to exert respective strengths and has a better effect compared with the single processing unit, data quantity of single transmission is increased after reinforcing communication interfaces, transmission times are reduced, and the integral acceleration effect is improved.

Description

technical field [0001] The invention relates to the field of biological information computing, in particular to an FPGA-based heterogeneous accelerated genetic computing method in a cloud platform. Background technique [0002] Parallel computing performed on heterogeneous computing systems is often referred to as heterogeneous computing. Heterogeneous computing has a wide range of applications, and almost all solutions to hugely challenging problems can be solved cost-effectively with heterogeneous computing. Typical applications include image understanding, particle tracing, acoustic beamforming, climate modeling, turbulent convective mixing simulations, and multimedia queries. These applications usually contain a variety of different computing types, so they are very suitable for solving with heterogeneous computing. [0003] The rapid growth of biological information and the continuous decline of data output costs will inevitably bring more data processing requirements...

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
IPC IPC(8): G06F19/00
CPCG16B50/00
Inventor 王新胜刘晓宁韩良罗敏
Owner HARBIN INST OF TECH AT WEIHAI
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products