Data distribution method and system for network data center

Abstract

A data distribution method of a network data center, comprising the following steps: at least one node server of the network data center downloads a source file from a data source server; the download management server stores the identification of the node server that has downloaded the source file; The node server that has not downloaded the source file queries the download management server for the identification of the node server that has downloaded the source file; the node server that has not downloaded the source file establishes a download connection with the node server that has downloaded the source file obtained through query, Downloading the source file from the node server that has downloaded the source file. In the above method, some node servers in the network data center download source files from the data source server, and other node servers download source files from node servers that have downloaded source files. It is not necessary for all node servers to download source files from the data source server, so that Reduce the use of leased line bandwidth between network data centers.

Description

【Technical field】 [0001] The invention relates to network technology, in particular to a data distribution method and system for a network data center. 【Background technique】 [0002] The Internet Data Center (IDC) is an Internet-based network base that includes multiple servers. Each server is a node of IDC for centralized collection, storage, processing and sending of data, so the server can also be called a node server. The node servers of the IDC are connected to each other to form a local area network. [0003] With the rapid development of the Internet, the business of Internet companies is also developing rapidly, and the number of IDC servers used to support the business is increasing day by day. An IDC deploys as many as tens of thousands or even hundreds of thousands of servers. In order to meet the needs of business expansion and take into account the factors of geographical distribution, multiple IDCs are often deployed in different regions to support services....

AI Technical Summary

Problems solved by technology

Since the communication between node servers of different IDCs needs to occupy the dedicated line bandwidth connected to the IDC, and the number of node servers in many IDCs is relatively large, and the update frequency of system software or application software is relatively fast, so the traditional method is between multiple IDCs. A large amount of private line bandwidth is required when distributing data between

[0006] In addition, in the traditional method, once an error occurs on the data source server, it may cause the node server that has not downloaded the source file to be unable to download the source file. Therefore, the traditional method has a low fault tolerance rate

Images