Resource Management

Memory Limits and Reservations

Setting Memory Limits

# Limit container to 512MB RAM
docker run -d --memory 512m myapp

# With swap (container can use 512MB + 256MB swap)
docker run -d --memory 512m --memory-swap 768m myapp

# No swap
docker run -d --memory 512m --memory-swap 512m myapp

# Memory reservation (soft limit)
docker run -d --memory 512m --memory-reservation 256m myapp

Memory Limits in docker-compose.yml

services:
  api:
    image: myapi
    deploy:
      resources:
        limits:
          memory: 512M
        reservations:
          memory: 256M

  db:
    image: postgres:16
    deploy:
      resources:
        limits:
          memory: 2G
        reservations:
          memory: 1G

Memory Flags Reference

OOM Killer

When a container exceeds its memory limit, Linux kills it:

# Check if container was OOM-killed
docker inspect --format '{{.State.OOMKilled}}' web
# true

# Disable OOM killer (use with caution!)
docker run -d --memory 512m --oom-kill-disable myapp

# Set OOM score adjustment (-1000 to 1000)
docker run -d --oom-score-adj -500 myapp

Monitor Memory Usage

# Real-time memory stats
docker stats --no-stream --format "table {{.Name}}\t{{.MemUsage}}\t{{.MemPerc}}"

# Check specific container
docker inspect --format '{{.HostConfig.Memory}}' web
# 536870912 (512MB in bytes)

# Check cgroup memory limit
cat /sys/fs/cgroup/memory/docker/<container_id>/memory.limit_in_bytes

Common Patterns

# Web server: 256MB-512MB
docker run -d --memory 512m nginx

# Node.js app: 512MB-1GB
docker run -d --memory 1g mynodeapp

# Java app: 70% of container memory for heap
docker run -d --memory 2g -e JAVA_OPTS="-Xmx1400m" myjavaapp

# Database: 4GB+
docker run -d --memory 4g postgres:16

CPU Limits and Shares

CPU Limits

# Limit to 1.5 CPU cores
docker run -d --cpus 1.5 myapp

# Limit to 50% of one core
docker run -d --cpus 0.5 myapp

# CPU shares (relative weight, default 1024)
docker run -d --cpu-shares 512 myapp   # Half the default

# Pin to specific CPUs
docker run -d --cpuset-cpus 0,1 myapp  # Use CPU 0 and 1

# Combine limits
docker run -d --cpus 2 --cpuset-cpus 0,1 myapp

CPU in docker-compose.yml

services:
  api:
    image: myapi
    deploy:
      resources:
        limits:
          cpus: "2.0"
          memory: 1G
        reservations:
          cpus: "0.5"
          memory: 256M

  worker:
    image: myworker
    deploy:
      resources:
        limits:
          cpus: "1.0"

CPU Flags Reference

I/O Limits

# Limit block I/O (bytes per second)
docker run -d --device-read-bps /dev/sda:10mb --device-write-bps /dev/sda:10mb myapp

# Limit I/O operations per second
docker run -d --device-read-iops /dev/sda:1000 --device-write-iops /dev/sda:1000 myapp

# Block I/O weight (10-1000, default 500)
docker run -d --blkio-weight 300 myapp

Network Bandwidth

Docker doesn't have built-in network bandwidth limiting. Use tc (traffic control) on the host:

# Limit container network to 10Mbit
tc qdisc add dev veth<container_id> root tbf rate 10mbit burst 32kbit latency 50ms

Resource Monitoring

# Real-time CPU and memory
docker stats --no-stream

# CPU usage over time
docker stats --format "{{.Name}} {{.CPUPerc}}" | while read line; do
  echo "$(date +%H:%M:%S) $line"
  sleep 1
done

# Find resource-hungry containers
docker stats --no-stream --format "table {{.Name}}\t{{.CPUPerc}}\t{{.MemUsage}}" | sort -k2 -h -r