Skip to main content

Documentation Index

Fetch the complete documentation index at: https://notes.kodekloud.com/llms.txt

Use this file to discover all available pages before exploring further.

In this lesson, you’ll learn how to manage memory and CPU usage in Docker containers. Proper resource constraints help ensure predictable performance and protect the host from runaway containers.

​
Table of Contents

​
Memory Limits

Docker containers run without memory constraints by default. You can view and modify these settings using docker run flags.

​
Default Behavior

Run a container named testone without any memory restrictions: 
docker container run -itd --name=testone ubuntu
Inspect its memory settings: 
docker container inspect testone | grep -i mem
Output: 
"Memory": 0,
"KernelMemory": 0,
"KernelMemoryTCP": 0,
"MemoryReservation": 0,
"MemorySwap": 0,
"MemorySwappiness": null
A value of 0 indicates “no limit.”

​
Hard Memory Limit

Set a hard cap on RAM usage with --memory. The following command limits the container to 200 MiB: 
docker container run -itd --name=testtwo --memory=200m ubuntu
Verify: 
docker container inspect testtwo | grep -i mem
Expected: 
"Memory": 209715200,
"KernelMemory": 0,
"KernelMemoryTCP": 0,
"MemoryReservation": 0,
"MemorySwap": 0,
"MemorySwappiness": null

​
Unlimited Swap

By default, swap is limited to the same size as --memory. To allow unlimited swap, use --memory-swap=-1: 
docker container run -itd \
  --name=testthree \
  --memory=200m \
  --memory-swap=-1 \
  ubuntu
Inspect: 
docker container inspect testthree | grep -i mem
Result: 
"Memory": 209715200,
"KernelMemory": 0,
"KernelMemoryTCP": 0,
"MemoryReservation": 0,
"MemorySwap": -1,
"MemorySwappiness": null
Unlimited swap (MemorySwap: -1) can lead to performance degradation if the host starts swapping heavily. Monitor your containers closely when using this option.

​
Soft Memory Reservation

With --memory-reservation, you set a soft limit that Docker tries to enforce under memory contention: 
docker container run -itd \
  --name=testfour \
  --memory=200m \
  --memory-reservation=100m \
  --memory-swap=-1 \
  ubuntu
Inspect: 
docker container inspect testfour | grep -i mem
Output: 
"Memory": 209715200,
"KernelMemory": 0,
"KernelMemoryTCP": 0,
"MemoryReservation": 104857600,
"MemorySwap": -1,
"MemorySwappiness": null
This configuration guarantees between 100 MiB and 200 MiB of RAM, with unlimited swap.

​
Memory Flags Summary

FlagDescriptionExample
—memoryHard limit for container memory--memory=200m
—memory-reservationSoft (guaranteed) memory reservation--memory-reservation=100m
—memory-swapTotal memory + swap limit (-1 = unlimited)--memory-swap=-1
—memory-swappinessTendency to swap (0–100)--memory-swappiness=10

​
CPU Limits

Docker lets you restrict CPU usage by setting the number of CPUs or pinning containers to specific cores.

​
Limiting CPU Count

First, check your host’s CPU count: 
nproc
Assuming 2 cores, limit the container to 1 CPU: 
docker container run -itd --name=testcpu --cpus=1 ubuntu
Inspect: 
docker container inspect testcpu | grep -i nano
You’ll see: 
"NanoCpus": 1000000000,
NanoCpus: 1000000000 represents 1 full CPU core.

​
Pinning to Specific Cores

Use --cpuset-cpus to bind a container to specific cores. For example, pin to core 1: 
docker container run -itd --name=testcpus --cpuset-cpus="1" ubuntu
Inspect: 
docker container inspect testcpus | grep -i cpuset
Output: 
"CpusetCpus": "1",
This container will only run on CPU core 1.

​
CPU Flags Summary

FlagDescriptionExample
—cpusNumber of CPU cores (fractional OK)--cpus=1.5
—cpuset-cpusBind container to specific cores--cpuset-cpus="0,1"

Watch Video