Seismic big data cluster parallel machine efficient sorting method and device

Abstract

The invention relates to the field of seismic exploration, in particular to an efficient sorting method and device for an earthquake big data cluster parallel machine, and the method comprises the steps: setting a data sorting task instruction, carrying out the primary cutting of preset earthquake data, and transmitting a primary data block to a calculation node; secondly, performing secondary segmentation on the first-level data block according to the number of idle cores in the computing node and the number of pre-distributed trace sets to form second-level data blocks, starting a plurality of thread sorting operations, inputting data by each thread according to the data blocks, then performing screening and re-beat on the data according to the trace sets, placing the data in a cache, and after the target data block is screened, performing re-beat on the data according to the trace sets to form the second-level data blocks; the same group of data in a plurality of caches is obtained to form a gather block, and the gather block is output to the designated position of the disk array, so that the large-scale seismic data is efficiently sorted through the distributed parallel data sorting model, computing resources are reasonably applied, the time cost of the data processing process is reduced, and the data sorting efficiency is effectively improved.

Description

technical field [0001] The invention relates to the field of seismic exploration, in particular to a high-efficiency sorting method and device for seismic big data cluster parallel machines. Background technique [0002] For oil exploration seismic data, in the process of data processing, it is necessary to rearrange the sequence of data volumes according to different key values ​​of trace heads, while the data rearrangement of massive large data volumes is conventionally done by extracting trace sets from Read all the target gathers one by one on the disk array. This kind of concentration of a large number of disk read and write requests requires a lot of I / O and computing resources. The efficiency is slow, and it is easy to cause read and write congestion, affecting the entire cluster. efficiency. [0003] For the above problems, no effective solution has been proposed yet. Contents of the invention [0004] The purpose of the present invention is to address the defect...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to address the defects and deficiencies of the prior art, to provide a method and device for high-efficiency sorting of seismic big data cluster parallel machines. By setting data sorting task instructions, the preset seismic data is cut into blocks once to form First-level data blocks, sending the first-level data blocks to computing nodes, and then performing secondary segmentation on the first-level data blocks according to the number of idle cores inside the computing nodes and the number of pre-allocated gathers to form two Level data blocks, start multiple threads for sorting operations, each thread implements data input by data block, and then screens and retakes data by channel, and places them in the cache. After the target data block is screened, it will be obtained The same group of data in several caches forms a gather, which is respectively output to the designated position of the disk array, realizing the efficient sorting of large-scale seismic data through a distributed and parallel data sorting model, and through the large-scale data sorting Input and output, solve the problem of reading and writing congestion in large-scale seismic data sorting, rationally apply computing resources, reduce the time cost of data processing, and effectively improve data loading efficiency

Images