To maximize the performance of IoT devices in edge computing, an adaptive polling technique that efficiently and accurately searches for the workload-optimized polling interval is required. In this paper, we propose NetAP-ML, which utilizes a machine learning technique to shrink the search space for finding an optimal polling interval. NetAP-ML is able to minimize the performance degradation in the search process and find a more accurate polling interval with the random forest regression algorithm. We implement and evaluate NetAP-ML in a Linux system. Our experimental setup consists of a various number of virtual machines (2-4) and threads (1-5). We demonstrate that NetAP-ML provides up to 23% higher bandwidth than the state-of-the-art technique.

This article investigates how to satisfy network performance requirements that are crucial in achieving the service level objectives (SLOs) in clouds. Traditional techniques for network performance management have a limited ability to satisfy the network SLOs. Our in-depth analysis reveals that the fundamental reason comes from decoupling of the CPU scheduler and the network traffic controller as the current CPU scheduler is not aware of such network requirements but only provides a fair-share amount of CPU to all containers. Thus, the container cannot perform the amount of network processing as needed to satisfy its SLO when the CPU allocation is insufficient. In this article, we propose Autothrottle that dynamically adjusts the CPU allocation for the containers to satisfy their network SLOs. The key element of Autothrottle is a throttle algorithm that autonomously determines the amount of CPU for each container needed to satisfy the requirement. We implement Autothrottle in the Linux kernel and evaluate it with massive real-world workloads such as Apache Kafka. Our evaluation results show that Autothrottle successfully satisfies the given network SLO only with a 2% gap while the existing scheme achieves 20% less than the SLO. We further observe that Autothrottle also reduces the CPU overhead in network processing by 19%, improving the network throughput by 27% compared to the existing scheme.