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

Heterogeneous parallel computing implementation method and device for three-dimensional sound wave NPML algorithm

A technology of parallel computing and implementation method, applied in the field of petroleum exploration, can solve the problems of poor power consumption efficiency and low computing time efficiency

Pending Publication Date: 2020-12-18
BC P INC CHINA NAT PETROLEUM CORP +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The embodiment of the present invention provides a method and device for implementing heterogeneous parallel computing of the three-dimensional acoustic wave NPML algorithm, which solves the technical problems of low computing time efficiency and poor power consumption efficiency caused by isomorphic computing in the prior art

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 parallel computing implementation method and device for three-dimensional sound wave NPML algorithm
  • Heterogeneous parallel computing implementation method and device for three-dimensional sound wave NPML algorithm
  • Heterogeneous parallel computing implementation method and device for three-dimensional sound wave NPML algorithm

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0037] Example 1, the configuration parameters of a customized super workstation of a certain brand are:

[0038] CPUs: Dual*Intel Xeon E5-2699v4 (22 cores / 44 threads)

[0039] Main memory: 256GB

[0040] Graphics card: Quadro-M6000 / 12GB

[0041] Accelerator card 1: Tesla-K80 / 24GB

[0042] Accelerator card 2: Radeon Instinct-MI25 / 16GB

[0043] Dual network card: 2*Gbe

[0044] Hard disk: 4*2TB

[0045] According to the above configuration parameters, you can create a figure 2 Computing resource topology shown.

example 2

[0046] Example 2, the configuration parameters of the first verification super workstation of the present invention are:

[0047] CPUs: Dual*Intel Xeon E5-2697v4 (18 cores / 36 threads)

[0048] Main memory: 128GB

[0049] Graphics card: Quadro-M5000 / 8GB

[0050] Accelerator card 1: Tesla-K40c / 12GB

[0051] Dual network card: 2*Gbe

[0052] Hard disk: 2*2TB

[0053] When detecting heterogeneous computing resources, the topological structure of its available computing resources is as follows: image 3 shown.

example 3

[0054]Example 3, the configuration parameters of the second verification super workstation of the present invention are:

[0055] CPUs: Dual*Intel Xeon E5-2697v2 (12 cores / 24 threads)

[0056] Main memory: 128GB

[0057] Graphics card: Quadro-K5000 / 4GB

[0058] Accelerator card 1: Tesla-K40c / 12GB

[0059] Dual network card: 2*Gbe

[0060] Hard disk: 2*2TB

[0061] When detecting heterogeneous computing resources, the topological structure of its available computing resources is as follows: Figure 4 shown.

[0062] Specifically, the main function of this step is to create a computing team with available computing devices in the topology of heterogeneous computing platforms, and provide basic conditions for subsequent computing task allocation. The present invention is a realization method of heterogeneous parallel computing based on the three-dimensional acoustic wave NPML algorithm. Therefore, when testing a computing device, the requirements of the algorithm must be ta...

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 provides a heterogeneous parallel computing implementation method and device for a three-dimensional sound wave NPML algorithm, and the method comprises the following steps: determiningavailable computing equipment in a known heterogeneous computing platform, constructing a computing resources topological structure according to the available computing equipment, wherein the available computing equipment in the heterogeneous computing platform comprises CPUs, GPUs and accelerators; according to the calculation scale of a single shot in the three-dimensional sound wave NPML algorithm, computing the memory of each available computing equipment in the resource topological structure, and allocating a corresponding computing task to each available computing equipment; according tothe computing resource topological structure, enabling each available computing equipment to allocate a corresponding computing task, achieving parallel computing of the three-dimensional sound waveNPML algorithm to acquire single-shot seismic simulation record data. According to the scheme, the total simulation calculation time of a single shot is shortened, the application timeliness is enhanced, the unit time efficiency is improved, and the maximum computing capability of the heterogeneous computing platform is exerted.

Description

technical field [0001] The invention relates to the technical field of petroleum exploration, in particular to a method and device for realizing heterogeneous parallel computing of a three-dimensional acoustic wave NPML algorithm. Background technique [0002] The 3D acoustic NPML algorithm forward modeling is an important tool for the analysis of the observation system in the design of the seismic exploration acquisition plan. Due to its large calculation workload and large amount of component data, in practical applications, the calculation time is too long and the intermediate data is too much. The requirement for computing resources is very high, which seriously affects the practical application. [0003] The realization of the existing three-dimensional acoustic wave NPML algorithm is to adopt the isomorphic computing method, that is, multi-machine CPU or multi-machine GP (Graphics Processing Unit, the processor of the graphics card is called a graphics processor). The...

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): G06F9/50
CPCG06F9/5016G06F9/5066
Inventor 黄兴贵皮红梅隆波高畅杨天福李平
Owner BC P INC CHINA NAT PETROLEUM CORP
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com