Method, apparatus for downloading data fragmentation by peer, and peer therefor

Abstract

The invention discloses a method for downloading data fragments by a peer, a device and the peer. At the first, the data fragments starting from a current playing position are divided into more than one aggregate according to time sequence, and the data segments needing download are determined by sequential selection of a first aggregate which starts from the current playing position and is positioned at the time sequence, if the first aggregate has the data fragments, the data fragments are selected to meet a real-time requirement, otherwise, a current peer is proved to have all data fragments contained in the aggregate and be capable of meeting the real-time requirement, and then the data fragments needing download are determined in a next aggregate in a least-number priority manner so as to meet a shared data maximization requirement, and after the data fragments needing download are determined, request information of downloading the data fragments is sent to a neighbor terminal, thus meeting the real-time requirement while meeting the shared data maximization requirement, which enhances the experience of a user.

Description

technical field [0001] The invention relates to the technical field of computer network communication, in particular to a method, a device and a peer applied to a peer downloading data slices in a P2P on-demand system. Background technique [0002] With the increasing development of network technology and the increasing popularity of network applications, the demand for data sharing between users is also becoming more and more significant, such as the sharing of simple data files, the sharing of data files during on-demand, etc. The way of data file sharing between peers is to divide the files to be downloaded into multiple equal-length data fragments, and each peer downloads the data fragments that it does not have but has from the neighbor end respectively. A way of data sharing, the data transmission process is carried out between the peer and its neighbors, without downloading from the server, reducing the dependence on the server. [0003] In this peer-to-peer download...

AI Technical Summary

Problems solved by technology

[0005] One of the methods of selecting data fragments in the prior art is to adopt a sequential selection method, that is, to select according to the sequence of each data fragment in the original data file, and only the data fragments before the current data fragment in the sequence The peer only selects the current data fragment when all the data fragments are already owned by the current peer. After the download of the current data fragment is completed, the next data fragment next to the current data fragment is selected. Fragmentation, this method of selecting data fragments is strictly in accordance with the order of selection. Therefore, in the on-demand system, for example, when playing multimedia files such as video files on demand, it can meet the needs of real-time and fluency. However, Since the selection is strictly in accordance with the sequence, each peer is more likely to have the same data fragmentation, so that the probability of requiring the other party's data fragmentation is reduced, and it is impossible to maximize the shared data between the peers

[0006] Another way to select data fragments in the prior art is to use the minimum optimization method, that is, to preferentially select the data fragment with the least number of shares among all neighbors. The data fragments that the peers do not have may be different from each other, so compared with the above order selection method, the probability of requiring the other party's data fragments increases, which meets the maximum requirement of sharing data between the end and the end, and After the minimum optimization method, the number of peers with the same data fragment increases, and accordingly the number of shares of each data fragment in peers and neighbors also increases, and the download speed also increases. It can be increased accordingly. However, in this way, for example, when a multimedia file is ordered in a P2P on-demand system, the data fragment with the least number of copies determined by the peer may not be located in the file playback sequence to a considerable extent. The data fragments are in the first order. Therefore, after the current data fragments are played, the next data fragments may not be downloaded from the neighbors. As a result, the on-demand delay and the playback process are not smooth, which cannot meet the real-time requirements. Require

Images