Method and system for selecting neighbor nodes for peer-to-peer network transmission

Abstract

The invention discloses a method for selecting neighbor nodes for peer-to-peer network transmission. The method comprises the following steps: establishing a connection between a central node and target nodes, which download target files, in a cluster, and judging whether the target nodes have non-downloaded file fragmentations or not; if YES, inquiring network usage amounts and total network card bandwidths of the target nodes; calculating residual bandwidths of the target nodes according to the network usage amounts and the total network card bandwidths of the target nodes, and sequencing the target nodes according to an order of the residual bandwidths from large to small to obtain a sequence table; selecting the previous N target nodes in the sequence table as the neighbor nodes. The method can increase the data distribution speed on the premise of reducing the influence on normal operation of tasks of the cluster. The invention further discloses a system for selecting the neighbor nodes for peer-to-peer network transmission, a computer readable storage medium and a server which have the above beneficial effects.

Description

technical field [0001] The invention relates to the field of data storage, in particular to a method and system for selecting neighboring nodes in peer-to-peer network transmission, a computer-readable storage medium and a server. Background technique [0002] A data center is a facility that accommodates computing resources, network connections, storage, and other related components. For stability and availability, it usually includes safety facilities such as redundancy, backup power, and air conditioning, and specific data center equipment is often configured very high. Data transmission between nodes is the most common problem in the data center, such as virtual machine image distribution, and the transfer of intermediate results during parallel computing. How to quickly transfer data between machines in the cluster has also become one of the hotspots in the data center. [0003] The BitTorrent (bit stream) protocol is widely used in the Internet for file exchange. A fil...

AI Technical Summary

Problems solved by technology

[0004] In the existing technology, tit-for-tat is applied to the data center environment, not only the network bandwidth is much larger than that of the Internet, but also all the machines in the cluster complete specific tasks under the unified scheduling of the administrator. While ensuring the speed of data distribution, sufficient bandwidth must be reserved for data exchange between machines during normal operation. For example, when running Spark distributed computing tasks, nodes also need bandwidth to ensure the exchange of intermediate results. Using the Internet based on The neighbor node selection algorithm of the tooth-for-tooth strategy, each node will try its best to upload the data in order to download the data as soon as possible, which will undoubtedly give up the cost and chase the end, which will greatly affect the tasks running on it

Images