Self-regulating dispatching method of streammedia server resources

Abstract

The invention relates to a streaming media server resource self-regulating scheduling method, which is composed of the following steps. Determine the position of each task of the streaming media task set in the ready queue using the RM algorithm, and set the activation condition to the lower priority task. The scheduler schedules according to the ready queue, transmits the media data and monitors the image quality. If the image quality is not satisfied, detects whether there is the CPU idle time between the finishing time of the current task and the beginning time of the next task. If there is, execute the image quality influenced task i. If not, check whether the activation condition of i task is satisfied. If it is, change the scheduling parameters of i task, adjust the streaming media data queue and reschedule. If the activation condition isn't satisfied, execute sequentially according to the original scheduling. The invention improves the streaming media service quality and the resource utilization ratio of the server.

Description

technical field [0001] The present invention relates to streaming media technology, in particular to a self-regulating scheduling method for streaming media server resources, which modifies the priority of tasks in real time according to streaming media service quality and resource utilization status, thereby improving streaming media service quality and server resources utilization rate. Background technique [0002] With the further popularization of the Internet, especially the development of broadband network technology, people's demand for multimedia information is increasing day by day, and a new multimedia broadband network technology - streaming media technology has emerged as the times require. Streaming media technology is a comprehensive concept technology, which includes video processing (video compression, decompression, encoding, decoding, etc.) technology, network transmission technology, scheduling technology, distributed processing technology and so on. Str...

AI Technical Summary

Problems solved by technology

Due to the large volume of streaming media data and fierce competition for resources, the use of traditional RM scheduling algorithms often results in some media tasks not being executed because they cannot compete for server resources, resulting in quality problems of streaming media services, such as image Blur, jitter, overlap, jump, pause, etc.

On the other hand, the RM scheduling algorithm is a static scheduling algorithm. In order to ensure the schedulability of tasks, some precious resources are always not effectively utilized and are wasted.

Images