When configuring a CentOS server with 16GB RAM for PHP/MySQL workloads, the SWAP allocation depends on several critical factors:

• Expected memory utilization patterns
• MySQL configuration and buffer pool settings
• PHP-FPM or mod_php memory limits
• System monitoring capabilities

The traditional "swap = RAM size" rule doesn't always apply to modern servers. For your 16GB Xeon system:

# Basic calculation based on Red Hat recommendations
if [ $RAM -le 2 ]; then
  SWAP=$((RAM * 2))
elif [ $RAM -le 8 ]; then
  SWAP=$((RAM * 1.5))
elif [ $RAM -le 16 ]; then
  SWAP=$RAM
else
  SWAP=16
fi

However, for a database-backed web server, consider these adjustments:

Your MySQL configuration significantly impacts SWAP needs. Check your current settings:

SHOW VARIABLES LIKE 'innodb_buffer_pool_size';
SHOW VARIABLES LIKE 'key_buffer_size';

Example configuration for 16GB server:

[mysqld]
innodb_buffer_pool_size = 8G
key_buffer_size = 256M
query_cache_size = 256M
tmp_table_size = 64M
max_heap_table_size = 64M

For PHP applications, implement proper memory controls:

; php.ini settings
memory_limit = 256M
max_execution_time = 120
realpath_cache_size = 256k
opcache.memory_consumption = 128

Set up monitoring to evaluate your actual needs:

# Create daily swap usage report
cat > /etc/cron.daily/swap_report < /var/log/swap_report
free -h >> /var/log/swap_report
echo "" >> /var/log/swap_report
vmstat 1 5 >> /var/log/swap_report
EOF
chmod +x /etc/cron.daily/swap_report

For your 16GB CentOS server running PHP/MySQL under medium-heavy load:

• Start with 8GB SWAP (50% of RAM)
• Monitor using vmstat and free
• Consider adding swap files if physical swap becomes insufficient:

# Add 4GB swap file if needed
fallocate -l 4G /swapfile2
chmod 600 /swapfile2
mkswap /swapfile2
swapon /swapfile2
echo '/swapfile2 none swap sw 0 0' >> /etc/fstab

Remember to adjust kernel swappiness for database servers:

echo 'vm.swappiness = 10' >> /etc/sysctl.conf
sysctl -p

html

When configuring a CentOS server with 16GB RAM running PHP/MySQL under medium-heavy load, SWAP allocation becomes critical for system stability. The traditional "double your RAM" rule is outdated for modern systems. Here's a data-driven approach:

Based on Red Hat's guidelines and real-world benchmarks:

• Minimum: 4GB (25% of RAM) - For systems with excellent memory monitoring
• Recommended: 8GB (50% of RAM) - Balanced approach for most production environments
• Conservative: 16GB (100% of RAM) - For systems running critical financial transactions

Here's how to create and optimize an 8GB SWAP file:

# Create swap file
sudo fallocate -l 8G /swapfile
sudo chmod 600 /swapfile
sudo mkswap /swapfile
sudo swapon /swapfile

# Make permanent
echo '/swapfile none swap sw 0 0' | sudo tee -a /etc/fstab

# Optimize swappiness (recommended: 10 for servers)
echo 'vm.swappiness=10' | sudo tee -a /etc/sysctl.conf
sudo sysctl -p

For MySQL servers, consider these my.cnf adjustments to reduce SWAP usage:

[mysqld]
innodb_buffer_pool_size = 12G  # 75% of 16GB
innodb_log_file_size = 2G
innodb_flush_method = O_DIRECT
query_cache_size = 0  # Disable for PHP workloads

Implement these monitoring solutions:

# Basic monitoring
free -h
vmstat 1

# Advanced tracking (install sysstat first)
sar -W 1 3  # Track swap in/out
sar -B 1 3  # Page statistics

Consider scaling up SWAP when:

• Regular OOM killer activity appears in /var/log/messages
• sar -B shows >5% page scanning
• MySQL experiences sudden performance drops during backups

For newer CentOS versions (7+), ZRAM can be more efficient:

# Install and configure
sudo yum install zram-generator
echo '[zram0]' | sudo tee /etc/systemd/zram-generator.conf
echo 'zram-size = min(ram, 8192)' | sudo tee -a /etc/systemd/zram-generator.conf
sudo systemctl daemon-reload
sudo systemctl start /dev/zram0