application optimization
First of all, from the application point of view, the best way to reduce CPU usage is of course to exclude all unnecessary work and keep only the core logic. For example, reduce the level of loops, reduce recursion, reduce dynamic memory allocation, and so on.
In addition, application performance optimization also includes many methods, the most common ones are:
-
Compiler optimization: Many compilers provide optimization options. If you enable them appropriately, you can get help from the compiler during the compilation phase to improve performance. For example, gcc provides the optimization option -O2, which automatically optimizes the code of the application after it is turned on.
-
Algorithm optimization: Using less complex algorithms can significantly speed up processing. For example, in the case of relatively large data, O(nlogn) sorting algorithms (such as quick sort, merge sort, etc.) can be used instead of O(n^2) sorting algorithms (such as bubble, insertion sort, etc.).
-
Asynchronous processing: Using asynchronous processing can prevent the program from being blocked because it is waiting for a certain resource, thereby improving the concurrent processing capability of the program. For example, by replacing polling with event notification, you can avoid the problem of polling consuming CPU.
-
Multi-thread instead of multi-process: Compared with the context switch of the process, the context switch of the thread does not switch the address space of the process, so the cost of the context switch can be reduced.
-
Make good use of cache: Frequently accessed data or steps in the calculation process can be cached in memory, so that they can be directly obtained from memory the next time they are used, speeding up the processing speed of the program.
System Optimization
From the perspective of the system, to optimize the operation of the CPU, on the one hand, it is necessary to make full use of the locality of the CPU cache to speed up cache access; on the other hand, it is to control the CPU usage of the process and reduce the mutual influence between the processes. Specifically, there are many CPU optimization methods at the system level:
-
CPU binding: Binding a process to one or more CPUs can improve the CPU cache hit rate and reduce context switching problems caused by cross-CPU scheduling.
-
CPU exclusive: Similar to CPU binding, CPUs are further grouped and processes are assigned to them through the CPU affinity mechanism. In this way, these CPUs are exclusively occupied by the specified process, in other words, no other process is allowed to use these CPUs.
-
Priority adjustment: use nice to adjust the priority of the process, the positive value lowers the priority, and the negative value increases the priority.
-
Set resource limits for processes: Use Linux cgroups to set the upper limit of CPU usage for processes, which can prevent system resources from being exhausted due to problems with an application itself.
-
NUMA (Non-Uniform Memory Access) optimization: A processor that supports NUMA will be divided into multiple nodes, and each node has its own local memory space. NUMA optimization is actually to let the CPU only access local memory as much as possible.
-
Interrupt load balancing: Regardless of whether it is a soft interrupt or a hard interrupt, their interrupt handlers may consume a lot of CPU. Enable the irqbalance service or configure smp_affinity to automatically load-balance the interrupt processing process to multiple CPUs.