Robust Security Mechnisms for Data Streams Systems
Mohamed Ali, Mohamed Y. Eltabakh, Cristina Nita-Rotaru · Purdue e-Pubs (Purdue University System) · 2004
Stream database systems are designed to support the fast on-line processing that characterizes many new emerging applications such as pervasive computing, sensor-based environments, on-line business processing and network monitoring. The sensitive nature of the data and the high-demands environment where data can be lost or dropped because of limited buffer storage or real-time constraints, require robust security mechanisms, i.e. mechanisms that not only provide security services, but are also fault-tolerant. In this paper we identify the security requirements for data stream systems, focusing on Nile a data stream management system. We present a new method, FT-RC4, that provides efficient and fault-tolerant data confidentiality. We demonstrate its applicability to data streams by using it as building block in the design of a security architecture for Nile and by presenting results for a stream based application.