Specifying and Managing Quality of Real-Time Data Services

Abstract

The demand for real-time data services is increasing in many applications including e-commerce, agile manufacturing, and telecommunication network management. In these applications, it is desirable to execute transactions within their deadlines, i.e., before the real-world status changes, using fresh (temporally consistent) data. However, it is challenging to meet these fundamental requirements due to dynamic workloads and data access patterns in these applications and potential conflicts between transaction timeliness and data freshness requirements. In this paper, we define average/transient deadline miss ratio and new data freshness metrics to specify the required quality of real-time data services. We also present a novel QoS management architecture to support the required QoS even in the presence of unpredictable workloads and access patterns. To prevent overload and support the required QoS, the presented architecture applies feedback control, admission control, and flexible freshness management schemes. A simulation study shows that our QoS-aware approach can support the required QoS, whereas baseline approaches fail to support the required miss ratio and/or freshness. Further, our approach shows a comparable performance to the theoretical oracle that is privileged by a complete future knowledge of data accesses.