Deep Analysis of Containerized Web Server Performance Using eBPF ‐Captured Data
Tomonori Takagaki, Kimihiro Mizutani · IEEJ Transactions on Electrical and Electronic Engineering · 2025
Focusing on containerized environments like Docker, we examine performance degradation in these environments using extended Berkeley Packet Filter (eBPF) technology. We develop a monitoring application with eBPF to analyze the relationship between system resources (CPU, memory, and disk I/O) and system calls. Using random forest regression to analyze a Web 3‐tier application, we found that the system calls writev, epoll_wait, and sched_yield significantly impact Web application performance, particularly in terms of CPU, memory, and disk I/O. © 2025 The Author(s). IEEJ Transactions on Electrical and Electronic Engineering published by Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.