Autonomous assurance technology for service continuity in demand-oriented information system

Abstract

In the progress of computing and network technologies, the amount of information in demand-oriented information services is being increased. Therefore, it is required to reduce the size of data and to provide continuous services. Additionally, users have heterogeneous service level requirements, but it is difficult for conventional centralized information system to meet the above needs simultaneously. Autonomous Decentralized Information System (ADIS) is proposed to achieve these requirements. In this system, the layered data is employed and each data is distributed on different levels. Therefore, the system can provide service with adaptability and minimum total storage volume. In this paper, ADIS is implemented as streaming video service and to satisfy the heterogeneous quality requirements, high and low service levels are shown for simplicity. As for the high service levels, the redundant hot-standby node mechanism is utilized, however, this is not provided for low service level because it may cause congestion in the whole system. To achieve continuous service for low service level without decreasing high's service level, the reservation backup approach is utilized. This is used for multiple users per slot to increase service continuity. Additionally, it is required to reserve nodes that are vacant within minimum steps as quick as possible. However, there is a trade-off between the node's congestion and the search area. To solve above problems, the autonomous assurance technology is proposed. In this technology, a mobile agent decides the scope of the search, concerning the congestion in the local area. Moreover, in the failure of a node, the recovery process will proceed automatically. The effectiveness of this technology is proved through implementation of practical streaming system.