Architecture and Task Scheduling of Video Streaming on Multi-core Platform

Abstract

Compared with traditional video streaming server, streaming on multi-core platform has many advantages: flexible and configurable on the number of executing core according to system requirements; fault-tolerant; and  fitting well to future process technologies, more cores will be available in advanced process technologies, meanwhile the complexity per core does not increase. In this paper, we focused on the video streaming issues on multi-core processor, including architecture and task scheduling. We proposed a pipeline-parallel hybrid multi-core architecture and service migration based task scheduling strategy on multi-core processor to improve the efficiency of video streaming and increase the number of concurrence. We implemented the task scheduling algorithm with proposed architecture, and provided evidences of 48% outperformance based on Cavium OCTEON CN5860 multi-core processor than full parallel architecture, meanwhile, request success rate higher than REM algorithm.

DOI : http://dx.doi.org/10.11591/telkomnika.v12i3.4491