Task tree-based large scale remote-sensing image parallel embedding method

Abstract

The invention relates to a task tree-based large scale remote-sensing image parallel embedding method, which comprising the following steps: 1) constructing an embedding task tree based on an abutting relation and a recursion task division, and constructing an embedding task tree according to the constructing method of a balance binary tree; 2) expressing a DAG (directed acyclic graph) model of the embedding task tree; 3) dynamically dispatching the embedding task tree based on CPDS-SQ (a dynamic DAG dispatching strategy based on a core path and a state queue); and 4) parallel processing a plurality of embedding tasks. The method has the following benefits: the data dependency relationship of a series of embedding tasks is decoupled from the MPI-based (message passing interface) parallel embedding realization process and is dispatched by the dynamic task tree for task dependency relationship control, so that the parallel control logic and parallel realization of the parallel embedding process can be greatly simplified, and the parallelism of large scale embedding is explored to the maximum extent to greatly improve the treatment performance and expansibility of large-scale embedding.

Description

Technical field [0001] The invention relates to a large-scale remote sensing image parallel mosaic method based on a task tree. Background technique [0002] Large-scale remote sensing image mosaics have been widely concerned and applied to scientific researches such as tropical rain forests, land use, and environmental changes in large areas or even the world. Remote sensing image mosaic usually seamlessly stitches a large amount of remote sensing image data with overlapping areas into a large geometrically accurate, radiation-balanced continuous mosaic, so as to provide a continuous and global view of the entire large geographic area. However, when the mosaic scale expands to a large area, the country, or even the world, large-scale mosaics face challenges: massive amounts of remote sensing image data, complex remote sensing image mosaic processing procedures, amazing computing power requirements, and a large number of precursors and subsequent data dependencies. MPI parallel ...

AI Technical Summary

Problems solved by technology

However, when facing the problem of large-scale remote sensing image mosaicking, purely relying on low-level parallel modes such as MPI or OpenMP will make the parallel processing process of large-scale remote sensing image mosaicing extremely complicated, and the extra tasks introduced by the dependencies between The communication and synchronization logic between computing nodes will be extremely complex and difficult

Images