Network distributed storage and reading method for file

Abstract

The invention discloses a network distributed storage and reading method for a file. The method comprises the following steps of: determining the storage disaster recovery parameter index (n, k, t and sigma) of a storage system (N1, N2,..., Nn); constructing a file subblock matrix Gn*k and a file aggregation matrix Hn*(n-k); equally dividing the file into k data blocks after the file is transformed into a binary format; carrying out information coding on the k data blocks of the file according to the file subblock matrix Gn*k, thereby obtaining n coded data blocks; and storing the n coded data blocks into n different storage nodes of the storage system. Compared with the prior art, the method has the advantages that the complete information of the file can be read and recovered through randomly selecting any n-t storage nodes from the n storage nodes of the storage system, namely, the condition that any t storage nodes of the system are simultaneously broken down can be tolerated, so that the disaster recovery capability of distributed data storage can be remarkably improved, and the dependence on data redundancy and equipment redundancy of the storage system is reduced.

Description

technical field [0001] The invention belongs to the technical field of coding organization and storage of electronic information files, and in particular relates to data disaster recovery and file reading technology for distributed storage of electronic information files on a network. Background technique [0002] With the rapid development of the network, it has become a trend for users to store or host data files on the network. However, the network covers a wide range, and some unpredictable accidental changes in any place may cause permanent loss or damage of some important data stored on the network, such as human operation errors, equipment component failures, earthquakes, floods, fires, typhoons , and even electronic warfare, which may cause immeasurable losses. The act of preventing such network data damage is called network data disaster recovery; in addition, due to the wide-area and unbalanced distribution of file data resources on the network, the existing netwo...

AI Technical Summary

Problems solved by technology

[0005] (2) When the scale of the network storage system expands, that is, when the total number of network storage nodes increases, the number of nodes with random failures and disasters will also increase by a percentage. If file replication technology is mainly used to prevent it, obviously there will be a large number of backup servers on the network In most of the time, at least one or more times the storage space should be idle, which will cause an astonishing waste of resources, and the larger the network size, the greater the waste;

[0006] (3) Even if the copies of each important file are limited to 2 to 5 copies and are stored in different physical storage nodes, as the distribution scale of network storage nodes expands, these 2 to 5 copies may be scattered in a large number of storage nodes among

This will highlight the problems of file resource location and file download path selection, and will restrict the service efficiency of large-scale network distributed file storage

Images