Configuration in software systems helps to ensure efficient operation and meet diverse user needs. Yet, some, if not all, configuration options have profound implications for the system's performance. Configuration performance analysis, wherein the key is to understand (or infer) the configuration options' relations and their impacts on performance, is crucial. Two major modalities exist that serve as the source information in the analysis: either the manual or source code. However, it remains unclear what roles they play in configuration performance analysis. Much work that relies on manuals claims their benefits of information richness and naturalness; while work that trusts the source code more prefers the structural information provided therein and criticizes the timeliness of manuals. To fill such a gap, in this paper, we conduct an extensive empirical study over 10 systems, covering 1,694 options, 106,798 words in the manual, and 22,859,552 lines-of-code for investigating the usefulness of manual and code in two important tasks of configuration performance analysis, namely performancesensitive options identification and the associated dependencies extraction. We reveal several new findings and insights, such as it is beneficial to fuse the manual and code modalities for both tasks; the current automated tools that rely on a single modality are far from being practically useful and generally remain incomparable to human analysis. All those pave the way for further advancing configuration performance analysis.

Virtual reality (VR) applications have revolutionized digital interaction by providing immersive experiences. 360$^{\circ }$VR video streaming has experienced significant growth and popularity as a pivotal VR application. However, the combination of limited network bandwidth and the demand for high-quality videos frequently hinders the achievement of a satisfactory quality of experience (QoE). Although prior methods have enhanced QoE, the effects of decoding latency have been poorly studied. It is technically challenging to design a quality adaptation algorithm that can balance the pursuit of high-quality videos and the limitation of limited bandwidth resources. To address this challenge, we propose an edge-end architecture for 360$^{\circ }$VR video streaming and aim to enhance overall QoE by solving a performance optimization problem. Specifically, our experiments on commercial mobile devices in real-world situations reveal that decoding latency significantly influences QoE. First, decoding latency plays a major role in contributing to end-to-end latency, which exceeds the transmission latency. Second, decoding latency can differ considerably between devices with varying computational capabilities. Building on this insight, we propose a novellatency-awarequalityadaptation (LAQA) algorithm. LAQA lies in developing a solution that can allocate video quality in real-time and enhance overall QoE. LAQA involves not only the quality of the received content, the transmission latency and the quality variance, but also the decoding latency and the fairness of the user quality. Subsequently, we formulate a combinatorial optimization problem to maximize overall QoE. Through extensive validation with experimental data from real-world situations, LAQA offers a promising approach to enhance QoE and ensure fairness performance in different devices. In particular, LAQA achieves 16.77% and 10.66% enhancement over the state-of-the-art combinatorial optimization and reinforcement learning algorithm, respectively, in terms of QoE at 4K resolution. Furthermore, LAQA ensures excellent scalability by simulating the number of users ranging from 15 to 60, making it a robust solution for diverse and growing user scales.

Gait recognition, i.e., recognizing persons from their walking postures, has found versatile applications in security check, health monitoring, and novel human-computer interaction. The millimeter-wave (mmWave) based gait recognition represents the most recent advance. Compared with traditional camera-based solutions, mmWave based gait recognition bears unique advantages of being still effective under non-line-of-sight scenarios, such as in black, weak light, or blockage conditions. Moreover, they are able to accomplish person identification while preserving privacy. Currently, there are only few works in mmWave gait recognition, since no public data set is available. In this paper, we build a first-of-its-kind mmWave gait data set, in which we collect gait of 95 volunteers 'seen' from two mmWave radars in two different scenarios, which together lasts about 30 hours. Using the data set, we propose a novel deep-learning driven mmWave gait recognition method called mmGaitNet, and compare it with five state-of-the-art algorithms. We find that mmGaitNet is able to achieve 90% accuracy for single-person scenarios, 88% accuracy for five co-existing persons, while the existing methods achieve less than 66% accuracy for both scenarios.