Fix High Load Average

Understanding Load Average

Load average shows processes waiting for CPU or I/O:

uptime
# 10:30:01 up 5 days, load average: 2.50, 1.80, 1.20

• 1-minute, 5-minute, 15-minute averages
• Load > number of CPU cores = overloaded
• Check cores: nproc or lscpu

Diagnose High Load

High load + low CPU usage = I/O wait:

top -c  # Look for %wa (I/O wait)
mpstat -P ALL 1

# Check disk I/O
iostat -x 1
sudo iotop -o

High load + high CPU = CPU-bound processes:

top  # Look for processes using >100% CPU

High Load Due to I/O Wait

Disk is bottleneck:

sudo iotop -o
sudo pidstat -d 1

sudo smartctl -a /dev/vda

Solutions for I/O wait:

• Move databases to faster storage (SSD NVMe)
• Reduce MySQL queries, add indexes
• Move logs to separate disk
• Use noatime mount option

High Load Due to CPU

htop -s PERCENT_CPU
ps aux --sort=-%cpu | head -10

• Optimize application code
• Add caching (Redis, Varnish)
• Upgrade VPS CPU
• Load balance across servers

High Load Due to Memory Pressure

vmstat 1
free -h

sudo fallocate -l 2G /swapfile
sudo chmod 600 /swapfile
sudo mkswap /swapfile
sudo swapon /swapfile

Process States

ps aux | awk '{print $8}' | sort | uniq -c

• R Running (CPU)
• D Uninterruptible sleep (I/O issue)
• S Sleeping (normal)
• Z Zombie process