摘要
现有的流媒体文件解析过程烦琐,对服务器性能要求高,解析的文件若存在错误易导致服务器运行异常甚至死机。因此提出了一种流媒体文件快速读取方法。主要特点是在流媒体服务器实时播放前对媒体文件解析打包,生成一种快速存取文件,下次有新用户时服务器会在磁盘上寻找对应的快速存取文件,用快速存取文件播放该媒体文件而不用通过线索样本重新生成RTP包,从而大量节省流媒体服务器读文件的处理时间,提高流媒体服务器的并发执行能力和系统的稳定性。实践证明,这种方法是简单的、可行的、高效的。
The existing parsing process for streaming media document is complicated and is strict at server. Furthermore it will lead to run unconventionally if the parsing document is wrong. The principal characteristic of double-quick accessing method for streaming media document proposed in this paper is to parse and pack video and audio data and bring out a double-quick storage document in streaming media server before playing. When a newcomer accesses server, server plays the media document by seeking the corresponding double-quick one in diskette instead of RTP packets recreated through hint sample. This would save vast disposal time of server and improve stability of system. Experiment results show that the benefit of this method is simple, feasible and high efficient.
出处
《微电子学与计算机》
CSCD
北大核心
2007年第11期114-115,119,共3页
Microelectronics & Computer
关键词
快速读取方法
实时传输协议
线索样本
轨迹
double-quick accessing method
transport protocol for real-time applications（RTP）
hint sample
track
