Implementation of Massive Real-time Database System Using Network Sensors and Sector Operation

Abstract: In order to improve the stability of the massive database system and its capabilities of data processing and services, this paper proposed a distributed information server cluster. The N-level distributed architecture based on the network sensors was used to form a massive database system for real-time collection and query. Due to support by the network sensors, all components in this architecture have the functionality of plug and play. To efficiently schedule the tasks of storing gathered data and querying information, a dynamic and self-adaptive scheduling algorithm based on task sensors was introduced in the application server. The task sensor collects the load status from the common processes in various information servers by information collection processes in the application server and sends them to the scheduler in the same server, which dispatches the tasks of data storing into the most appropriate information server. Furthermore, an appropriative database system based on sector read-write directly was presented to improve the access speed, which is almost 25 times than that of database based on SYBASE or ORACLE. The practice shows that the system developed by this strategy has good flexibility an