Docker restricts the use of CPU by containers

In the previous section, you learned how to limit the use of memory by containers. In this section, we take a look at CPU.

By default, all containers can use host CPU resources equally and without restrictions.

Docker can set the weight of CPU used by the container with-c or-- cpu-shares. If not specified, the default value is 1024.

Unlike the memory limit, the cpu share set with-c is not the absolute number of CPU resources, but a relative weight value. The CPU resources that a container can eventually be allocated depend on its cpu share as a percentage of the total cpu share of all containers.

In other words: through cpu share, you can set the priority for the container to use CPU.

For example, two containers are launched in host:

Docker run-name "container_A"-c 1024 ubuntudocker run-name "container_B"-c 512 ubuntu

Container_A 's cpu share 1024 is twice that of container_B. When both containers require CPU resources, container_A can get twice as much CPU as container_B.

It is important to note that this weighted CPU only occurs when CPU resources are tight. If container_A is idle, container_B can also be allocated to all available CPU in order to make full use of CPU resources.

Let's continue to experiment with progrium/stress.

1. The startup container_A,cpu share is 1024:

Cpu is used to set the number of worker threads. Because there is only one CPU in the current host, a worker thread can fill up the CPU. If the host has more than one CPU, you need to increase the number of cpu accordingly.

2. The startup container_B,cpu share is 512:

3. Execute top in host to check the container's use of CPU:

Container_A consumes twice as much CPU as container_B.

4. Pause container_A now:

5. Top shows that container_B can use the whole CPU when container_A is idle:

CPU limit is discussed here, the above is the whole content of this article, I hope it will be helpful to your study, and I also hope that you will support it.