Quick analysis and storage method of big data based on database

Abstract

A quick analysis and storage method of big data based on a database includes the steps 1, according to data content actually to be analyzed, defining a data format to a file; 2, reading the data format from the file to a data structure constructed in a memory, as data analysis basis; 3, making preparation before data analysis; 4, analyzing data; 5, storing the data; 6, after a list in the step 5 finishes storage, clearing a results list, setting a processing completion mark to be false, setting a status of the processing completion mark to be idle, recovering the processing completion mark to a thread pool, and waiting for allocation of new data blocks. The method has the advantages that analytical speed is high, the analyzed data structure is configurable and highly universal, data storage redundancy is low, and stored results data are convenient to analyze afterwards.

Description

technical field [0001] The invention relates to a computer data processing technology, and more specifically, relates to solving the problems of fast analysis and post-event analysis of big data. Background technique [0002] In many projects, a large amount of data that needs to be processed at high speed will be encountered. If these data cannot be processed in time, it will have a fatal impact on the entire software system. The existing big data analysis solutions basically deal with fixed formats. Yes, the scalability compatibility is very poor, and the processing speed on a single computer is not fast enough, and the existing technology is not fully utilized to maximize the processing capacity of the computer. Distributed computing requires high network speed and stability, and An efficient and comprehensive data storage method is not provided, so the extraction of data after processing is not flexible. [0003] After searching, the application number is 200810097594.7...

AI Technical Summary

Problems solved by technology

[0004] 1. Versatility: The files in the above patents need to set the data format, which has poor scalability and versatility;

[0005] 2. Processing efficiency: A large part of the design logic in the above patents is how to split and combine files, and the corresponding disk reading and writing and network transmission will consume a lot of time and resources;

[0006] 3. Data integrity: the above-mentioned patent adopts the server-client model, which requires distributed computing, but the processing of data by a single client does not make full use of computer resources, and if the network connection is abnormal, it may cause unpredictable time delays or data loss with serious consequences

Images