Autonomous Data Reallocation Technology for High Assurance in Streaming Service System

Abstract

With the advances of compression, storage and network technologies, online streaming service is becoming more and more popular. However, it is difficult for conventional systems to meet the continuous and heterogeneous requirements from service providers and users simultaneously. This paper introduces an autonomous decentralized VoD system to provide service in heterogeneous quality with adaptability and fewer additional storage volume. In this system, service provider allocates layered streaming data in separated nodes distributed in different levels. The users send pull mobile agents to select appropriate nodes for required service quality. However, for changing preference of users and local congestion, the response time will become worse. To increase the capacity of the edge level with base layered data, Autonomous Data Reallocation Technology is proposed by dynamically reallocating the layered data between the nodes in different levels. Moreover, the tradeoff between Service Level Agreement (SLA) and response time is analyzed. The effectiveness of the proposed technology is proved by comparison with conventional system.