Introduction

Key-value stores provide a distributed solution to cloud computing and big data management. Generally, key-value stores employ a weak consistency model, which relaxes the data consistency to improve system performance on query response. However, a drawback is that data accessed by users might be stale. Hence, there is an intrinsic trade-off between query latency and data consistency. At a local node, data consistency is expressed as the freshness of data accessed by queries. Hence, the trade-off at node level boils down to finding a suitable trade-off between query latency (i.e., quality of service (QoS)) and data freshness (i.e., quality of data (QoD)). This chapter provides an introduction of quality-aware scheduling for key-value stores which balances the aforementioned trade-off. In the following, Sect. 1.1 introduces the application scenarios of quality-aware scheduling for key-value data stores; Sect. 1.2 highlights the significance and challenges of the research in this book; Sect. 1.3 illustrates an implementation framework of this study; Sect. 1.4 provides an overview of this book.