CPU limits and reservations in Docker - Step-by-Step Execution

Process Flow - CPU limits and reservations

Start Container

↓

Check CPU Reservation

↓

Guarantee CPU Shares

↓

Check CPU Limit

↓

Restrict CPU Usage

↓

Run Container with Constraints

This flow shows how Docker starts a container by first reserving CPU shares, then applying CPU usage limits, and finally running the container with these CPU constraints.

Execution Sample

Docker

docker run --cpus=1.5 --cpu-shares=512 alpine top

Runs an Alpine container limiting CPU usage to 1.5 CPUs and reserving CPU shares of 512.

Process Table

Step	Action	Parameter Checked	Value	Effect on Container
1	Start container creation	-	-	Prepare container environment
2	Check CPU reservation	cpu-shares	512	Container guaranteed relative CPU priority
3	Check CPU limit	--cpus	1.5	Container limited to max 1.5 CPUs usage
4	Apply CPU constraints	cpu-shares & --cpus	512 & 1.5	CPU scheduler enforces limits
5	Run container	-	-	Container runs with CPU limits and reservations
6	Exit	-	-	Container runs until stopped or exited

💡 Container runs with CPU limits and reservations applied as specified

Status Tracker

Variable	Start	After Step 2	After Step 3	Final
cpu-shares	default 1024	512	512	512
cpu-limit (cpus)	unlimited	unlimited	1.5	1.5

Key Moments - 3 Insights

Why does setting --cpu-shares not limit CPU usage strictly?

Because cpu-shares only set relative priority for CPU time when CPU is contended, not a hard limit. See execution_table step 2 vs step 3.

What happens if --cpus is not set but cpu-shares is?

The container can use unlimited CPU but with relative priority defined by cpu-shares. This is shown by cpu-limit variable staying 'unlimited' in variable_tracker after step 2.

Can CPU usage exceed the --cpus limit?

No, --cpus sets a hard limit enforced by the scheduler, so usage cannot exceed this. See execution_table step 3 and 4.

Visual Quiz - 3 Questions

Look at the execution_table at step 3, what CPU limit value is applied?

A Unlimited CPU

B 512 CPU shares

C 1.5 CPUs

D 2 CPUs

💡 Hint

Check the 'Value' column at step 3 in execution_table

According to variable_tracker, what is the cpu-shares value after step 2?

A 512

B 1024

C 2048

D 256

💡 Hint

Look at the 'cpu-shares' row under 'After Step 2' in variable_tracker

If --cpus was removed from the command, how would the cpu-limit variable change in variable_tracker?

A It would become 512

B It would remain 'unlimited'

C It would become 1.5

D It would become 0

💡 Hint

Refer to variable_tracker 'cpu-limit' value before and after step 3

Concept Snapshot

Docker CPU limits and reservations:
- --cpu-shares sets relative CPU priority (default 1024)
- --cpus sets hard CPU usage limit (e.g., 1.5 CPUs)
- cpu-shares affect scheduling when CPU is busy
- --cpus restricts max CPU usage strictly
- Combine both for guaranteed and limited CPU usage

Full Transcript

This visual execution shows how Docker applies CPU limits and reservations when running a container. First, the container is prepared. Then Docker checks and applies CPU shares, which set the relative priority for CPU time when the CPU is busy. Next, Docker applies the CPU limit using the --cpus option, which restricts the container's maximum CPU usage strictly. Both settings are enforced by the CPU scheduler before the container runs. The variable tracker shows cpu-shares changing from the default 1024 to 512, and cpu-limit changing from unlimited to 1.5 CPUs. Key moments clarify that cpu-shares do not limit CPU usage strictly but prioritize it, and --cpus sets a hard limit. The quizzes test understanding of these values at different steps.