Linux kernel tuning11/13/2023 If they do, calculate the smp_affinity mask and set the corresponding configuration file.If they do not, all interrupts of these devices are routed to the same CPU and cannot be modified.For other devices, check the chip manual to see whether these devices support distributing interrupts.You can no longer manually configure interrupt affinity for such devices. Starting from CentOS 7.5, the system automatically configures the best interrupt affinity for certain devices and their drivers, such as devices that use the be2iscsi driver and NVMe settings. Identify the devices that need to balance interrupts.Automatic balance can be implemented through the irqbalance service. The command for this operation is cpupower frequency-set -governor performance. To ensure service performance, select the performance mode and fix the CPU frequency at the highest supported operating frequency without dynamic adjustment. CPU-frequency scalingĬpufreq is a module that dynamically adjusts the CPU frequency. This section introduces performance tuning based on the classified kernel subsystems. For detailed usage of the BCC toolkit, see BPF Compiler Collection (BCC). Compared with kprobe, BPF provides higher security and is more suitable for the production environments. Compared with perf/ftrace, BPF provides programmability and smaller performance overhead. Therefore, you can choose proper tools to conduct an in-depth analysis based on the results in In 60 seconds. Starting from CentOS 7.6, the Linux kernel has supported Berkeley Packet Filter (BPF). Perf is an important performance analysis tool provided by the Linux kernel, which covers hardware level (CPU/PMU, performance monitoring unit) features and software features (software counters, trace points). You can analyze outputs of the following list items to troubleshoot most common performance issues.įor detailed usage, see the corresponding man instructions. All tools used can be obtained from the official release of Linux. Linux Performance Analysis in 60,000 Milliseconds is published by the author Brendan Gregg and the Netflix Performance Engineering team. This section lists common methods for performance analysis. System tuning must be based on the results of system performance analysis.
0 Comments
Leave a Reply.AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |