Seismic processing multi-core heterogeneous parallel mode adaptive matching general computation module

An adaptive matching, general-purpose computing technology, applied in multi-programming devices, program control devices, etc., can solve the problems of high complexity of programming language development and inability to automatically adapt, and achieves simplified coding workload and good ease of use , the effect of high code reusability

Inactive Publication Date: 2015-04-29
CHINA PETROLEUM & CHEM CORP +1
View PDF4 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the development complexity of the programming languages ​​corresponding to these parallel modes is high, and there are different characteristics. Developers need to have certain parallel development experience an

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
  • Seismic processing multi-core heterogeneous parallel mode adaptive matching general computation module
  • Seismic processing multi-core heterogeneous parallel mode adaptive matching general computation module
  • Seismic processing multi-core heterogeneous parallel mode adaptive matching general computation module

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] Example 1. A universal computing model for adaptive matching of multi-core heterogeneous parallel modes for seismic processing is characterized in that: the method for constructing the computing model includes the following steps:

[0032] (1) By analyzing the target problem of multi-core heterogeneous parallel computing, divide multi-core heterogeneous parallel tasks;

[0033] (2) Establish a multi-core and heterogeneous general parameter library based on the parameters of multi-core and heterogeneous machines;

[0034] (3) Establish a multi-core heterogeneous parallel algorithm library based on parallel tasks and general parameter libraries;

[0035] (4) Design an interpretation system that can adaptively select corresponding parallel algorithms in the multi-core heterogeneous parallel algorithm library according to different multi-core heterogeneous software and hardware parameters;

[0036] (5) Developers realize corresponding parallel algorithm programming for different mul...

Embodiment 2

[0047] The pre-stack depth migration of the wave equation of the shot field has great application value in the seismic data processing process of oil and gas exploration and development, especially for the structural imaging of oil and gas exploration and development in complex structural areas and its subsequent lithology processing, attribute processing and geology Explanations have very important practical significance. The following takes the wave equation pre-stack depth migration as an example to introduce the use effect of the present invention suitable for the general model of seismic processing multi-core heterogeneous parallel computing. Assume that the current computing system contains cluster, CPU multi-core, and GPU multi-core computing resources. The cluster is configured with MPI computing environment, CPU multi-core configuration OpenMP computing environment, GPU multi-core configuration CUDA operating environment. Analyze the serial processing flow of wave equ...

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 seismic processing multi-core heterogeneous parallel mode adaptive matching general computation module. According to the model, a corresponding multi-core heterogeneous parallel algorithm is designed by analyzing the multi-core heterogeneous parallel calculation target problem; the parallel algorithm is interpreted into a parallel function based on different software and hardware architectures through an interpreting system, in addition, a software and hardware general parameter base of the current system is referred, and the system processing efficiency is improved; by aiming at different multi-core heterogeneous software and hardware interfaces, developers realize corresponding parallel program programming; parallel programs are compiled into machine languages through a corresponding compiler, and are operated on the corresponding multi-core heterogeneous software and hardware architectures. The computation module provided by the invention has the advantages that the development difficulty of the developers is simplified, and in addition, the portability and the expandability of the model are improved through expanding and updating a multi-core heterogeneous general parameter base and an algorithm base.

Description

Technical field [0001] The invention belongs to the field of high-performance computing for seismic data processing, and specifically relates to a universal computing model for adaptive matching of multi-core heterogeneous parallel modes for seismic processing. Background technique [0002] In the field of general parallel computing, more and more parallel computing modes and parallel frameworks have emerged. For example, MPI, MAP / REDUCE in clusters, OpenMP and OpenACC in multi-core CPUs, CUDA and OpenCL in multi-core GPUs, and HMPP in general frameworks. However, the programming languages ​​corresponding to these parallel modes have high development complexity and different characteristics. Developers need to have certain parallel development experience and be proficient in using different types of programming environments. When the software and hardware architecture in the system changes, Developers also need to modify the original program and cannot adapt automatically. Summ...

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): G06F9/46G06F9/44
Inventor 孟祥宾王延光单联瑜隋志强
Owner CHINA PETROLEUM & CHEM CORP
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