Key points are not available for this paper at this time.
Timeout is commonly used to handle unexpected failures in server systems. However, improper use of timeout can cause server systems to hang or experience performance degradation. In this paper, we present TScope, an automatic timeout bug identification tool for server systems. TScope leverages kernel-level system call tracing and machine learning based anomaly detection and feature extraction schemes to achieve timeout bug identification. TScope introduces a unique system call selection scheme to achieve higher accuracy than existing generic performance bug detection tools. We have implemented a prototype of TScope and conducted extensive experiments using 19 real-world server performance bugs, including 12 timeout bugs and 7 non-timeout performance bugs. The experimental results show that TScope correctly classifies 18 out of 19 bugs. Compared to existing runtime bug detection schemes, TScope reduces the average false positive rate from 47.24% to 0.8%. TScope is light-weight and does not require application instrumentation, which makes it practical for production server performance bug identification.
He et al. (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: