Real-time stream processing is essential for many real-life stream-based applications. Systems designed to run such applications must be prepared to operate under overloaded conditions. In this paper, the load shedding problem is studied for an important class of real-time data stream monitoring applications. In particular, we adopt the (Formula presented.) deadline model, instead of the commonly used random dropping policy, to capture the QoS requirements of such applications. Based on this model, we propose a Safe lOad Shedding Approach (SOSA) that aims to reduce the workload imposed on the system while at the same time preserve system timing constraints by exploiting data stream semantics. SOSA categorizes stream processing into two different modes and allows one to place provably lighter loads on streams that operate in one particular mode. To demonstrate the usefulness of SOSA, we introduce a concrete (Formula presented.) scheduling algorithm called SOSA-DBP by combining SOSA with DBP, a well-known (Formula presented.) scheduling algorithm. Probabilistic analysis and experimental results show that SOSA-DBP has significant performance gain over DBP.
All Science Journal Classification (ASJC) codes