Skip to main content

Chapter 12.2 - Performance Tuning

Learning Objectives​

By the end of this chapter, you will be able to:

  • Identify bottlenecks
  • Optimize CPU performance
  • Optimize memory
  • Optimize disk I/O
  • Configure kernel parameters
  • Monitor performance

Identifying Issues​

Diagnostic Tools​

# CPU
top, htop, mpstat
cat /proc/stat

# Memory
free -h
vmstat
cat /proc/meminfo

# Disk
iostat -x
iotop
df -h

# Network
iftop, nethogs
ss -i

Key Metrics​

# Load average
uptime
# Interpretation: < number of CPUs

# CPU idle
mpstat 1
# Goal: > 20% idle

# Available memory
free -h
# Goal: > 10% free

# I/O wait
iostat -x 1
# Goal: < 5%

CPU Optimization​

Process Priorities​

# Adjust nice values
nice -n 10 command  # Low priority
nice -n -10 command  # High priority (root)

# Renice existing process
renice 10 PID
renice 10 -u user1

CPU Affinity​

# Assign process to specific CPU
taskset -c 0,1 command

# Existing process
taskset -cp 0,1 PID

CPU Limits​

# cgroups v2
systemd-run --slice=my-slice --property=CPUQuota=50% command

# cpuset
cset set -c 0-3 -s my-set
cset proc -m -p PID -s my-set

Memory Optimization​

Swap​

# View swap
free -h
swapon --show

# Adjust swappiness
# /etc/sysctl.conf
vm.swappiness=10  # 0-100, lower = less swap

# Apply
sudo sysctl -p

Cache and Buffer​

# Clear cache (if necessary)
sync
echo 3 | sudo tee /proc/sys/vm/drop_caches

# Check usage
free -h

Memory Limits​

# ulimit
ulimit -v 1000000  # 1GB in KB

# cgroups
systemd-run --slice=my-slice --property=MemoryLimit=1G command

I/O Optimization​

File Systems​

# Choose appropriate FS
# - ext4: General
# - xfs: Large files
# - btrfs: Snapshots

# Mount options
# /etc/fstab
/dev/sda1  /data  xfs  defaults,noatime  0 2
# noatime: No atime update (performance)

I/O Scheduler​

# View current scheduler
cat /sys/block/sda/queue/scheduler

# Change scheduler
echo deadline > /sys/block/sda/queue/scheduler

# Permanent: kernel parameter
# elevator=deadline

RAID and LVM​

# RAID for performance
# - RAID 0: Performance (no redundancy)
# - RAID 10: Performance + redundancy

# LVM for flexibility
# - Volume extension
# - Snapshots

Kernel Parameters​

sysctl​

# View parameters
sysctl -a

# Modify temporarily
sudo sysctl vm.swappiness=10

# Permanent: /etc/sysctl.conf
vm.swappiness=10
net.core.rmem_max=16777216
net.core.wmem_max=16777216

# Apply
sudo sysctl -p

Network Parameters​

# /etc/sysctl.conf
# TCP optimization
net.core.somaxconn=1024
net.ipv4.tcp_max_syn_backlog=2048
net.ipv4.tcp_fin_timeout=30
net.ipv4.tcp_keepalive_time=300

Memory Parameters​

# /etc/sysctl.conf
vm.swappiness=10
vm.dirty_ratio=15
vm.dirty_background_ratio=5

Service Optimization​

Unnecessary Services​

# View active services
systemctl list-units --type=service --state=running

# Disable unnecessary services
sudo systemctl disable service-name
sudo systemctl stop service-name

Resource Limits​

# Unit file systemd
[Service]
MemoryLimit=512M
CPUQuota=50%
IOWeight=100

Performance Monitoring​

Monitoring Scripts​

#!/bin/bash
# Monitoring performance

echo "=== CPU ==="
top -bn1 | grep "Cpu(s)"
mpstat 1 1

echo "=== Memory ==="
free -h
vmstat -s

echo "=== Disk ==="
iostat -x 1 1
df -h

echo "=== Network ==="
ss -i

Performance Alerts​

#!/bin/bash
# Performance alerts

# CPU
CPU_IDLE=$(mpstat 1 1 | tail -1 | awk '{print $NF}')
if (( $(echo "$CPU_IDLE < 20" | bc -l) )); then
    echo "ALERT: CPU idle at ${CPU_IDLE}%"
fi

# Memory
MEM_FREE=$(free | grep Mem | awk '{printf "%.0f", $4/$2 * 100}')
if [ "$MEM_FREE" -lt 10 ]; then
    echo "ALERT: Memory free at ${MEM_FREE}%"
fi

Summary​

In this chapter, you learned:

Diagnostic: Tools for identifying bottlenecks
CPU: Priorities, affinity, limits
Memory: Swap, cache, limits
I/O: FS, scheduler, RAID
Kernel: sysctl, network/memory parameters
Services: Disable unnecessary, limits
Monitoring: Scripts, performance alerts

Next Steps​

Chapter 12.3: Complete Automation
End of Course: Recap

Chapter created on: December 2024