为精确地分析和评估对等(P2P)算法的性能,在并行分布式网络模拟器的基础上,设计并实现数据包级的通用可扩展P2P算法模拟器.该模拟器可实现达万级节点的大规模P2P算法仿真.模拟器采用分层框架结构,包括P2P用户模拟层、P2P算法模拟层、P2P传输模拟层和并行分布式连接层,并完成对传输层UDP和TCP套接字扩展、改进,将P2P应用层算法控制消息解析与网络层数据分片传输相结合,支持大规模P2P实时和非实时算法.结合BitTorrent阐明本模拟器的设计与实现,达到万级节点的模拟实验结果验证本模拟器的正确性和有效性.
In order rithm, a universal packet-level. This to actually analyze and evaluate performance of large-scale peer-to-peer (P2P) algo- and scalable simulation P2P platform was designed and implemented in terms of platform is able to support large-scale simulation of P2P algorithms with the scaleof 10 thousand nodes. Hierarchical framework was P2P algorithm layer and P2P transmission layer as adopted in the platform, including P2P user layer, well as parallel distributed connection layer. It ira- proves UDP and TCP socket which are suitable for parallel distributed network simulator (PDNS)-based P2P platform. In addition, it combines the analysis of control messages in terms of P2P applica- tion layer and data transmission by fragmentation in terms of network layer. Thus, the platform can support both real-time and non real-time P2P algorithms. Taking BitTorrent as instance, the detailed design and implementation of the platform was described. Finally, the correctness and effectiveness were proved by the simulation results up to the scale of 10 thousand nodes.