The cloud-edge collaborative edge computing paradigm is driving the comprehensive development of the Internet of Things. With the widespread deployment of 5G mobile networks, IoT end devices tend to offload data to more powerful edge nodes for computing, thereby improving the speed. This has led to an explosive growth in the amount of data processed by edge nodes. General CPU scheduling strategies are unsuitable for edge nodes that frequently occur in data-intensive applications. In this study, we focus on application feature awareness and design a strategy that considers data arrival rate and cost-fair CPU scheduling (RCFS) with two components: CPU resource allocation based on process weights and process scheduling based on distributed weighted round-robin. Compared to the Linux default scheduling strategy, CFS, experiments on edge nodes show that the RCFS strategy can effectively improve CPU utilization, decrease the running time of data-intensive applications, and improve the system's data throughput. In the best-case scenario, the RCFS strategy enhanced CPU utilization by up to 98.5%, decreased running time by up to 45.5%, and increased data throughput by up to 58.7%.