Towards Query Latency Optimization in the Fog

The streaming data is increasing at an exceeding rate due to the proliferation of IoT (Internet of Things)-enabled applications. Many IoT applications demand fast analytics of streaming data for real-time responses. Thus, it is crucial to have an effective query optimization technique. The query optimization within fog is a challenging issue due to its heterogeneity and being a highly resource-constrained environment. In this paper, we propose a query optimization approach that optimally performs both the data approximation and query task execution in the fog to drastically reduce query latency with efficient resource utilization. We introduce a novel optimal fog node selection algorithm which performs the optimal set selection based on minimum-cost criteria while considering the current network conditions. Additionally, we present an efficient compute congestion-aware scheduling technique that performs task scheduling by taking into account the resource capacity of fog nodes. Our evaluation results demonstrate that the proposed approach outperforms other techniques in terms of latency and bandwidth consumption.