Optimize Memory Usage with zram and Swap

Managing memory efficiently is crucial for system stability and performance. Modern Linux systems provide versatile tools such as zram and traditional swap to extend physical RAM capabilities. This article presents a detailed guide to configuring, tuning, and monitoring zram and swap, along with best practices and security considerations.

1. Understanding zram and Swap

1.1 zram Key Characteristics

• Compression: Reduces memory footprint by 2–4× depending on data.
• Speed: Faster than disk swap since it uses RAM and CPU compression.
• Use Case: Ideal for devices with limited RAM (IoT, single-board computers).

1.2 Swap Key Characteristics

• Persistence: Survives reboots when using a swap partition or file.
• Capacity: Limited only by disk size, but slower due to I/O latencies.
• Use Case: Systems with abundant storage but occasional memory peaks.

2. zram vs. Swap: Comparison Table

3. Configuring zram

1. Install zram tools:
   Debian/Ubuntu: apt install zram-tools
   RHEL/CentOS: yum install zram
2. Define zram devices:
   sudo modprobe zram num_devices=1
3. Allocate size:
   echo 2G sudo tee /sys/block/zram0/disksize
4. Format and enable:
   sudo mkswap /dev/zram0 sudo swapon /dev/zram0

3.1 Persistent Configuration

Create /etc/systemd/zram-generator.conf:

[zram0]
zram-size = min(ram, 4096) # Limit to 4G or RAM size

4. Configuring Swap

• Partition-based swap: Use fdisk or parted to create a swap partition, then mkswap and swapon.
• Swap file:
  fallocate -l 4G /swapfile
chmod 600 /swapfile
mkswap /swapfile swapon /swapfile
• Fstab entry:
  /swapfile none swap sw 0 0

5. Tuning Swappiness and Compression

Linux uses the swappiness parameter to determine swap usage aggressiveness. Lower values favor RAM, higher values allow more swapping.

• Check current value: cat /proc/sys/vm/swappiness
• Temporarily set: sysctl vm.swappiness=10
• Persist in /etc/sysctl.conf:

vm.swappiness=10

Adjust zram compression algorithm:

echo lzo >/sys/block/zram0/comp_algorithm

6. Monitoring and Benchmarks

• htop or top for overall memory usage.
• zramctl to inspect zram devices.
• swapon -s to list active swap.
• Run stress-ng or real workloads to measure responsiveness and throughput.

7. Best Practices

• Combine zram and swap: Use zram for short-term, high-speed compression, and disk swap as a fallback for extreme memory pressure.
• Right-size: Allocate zram equal to 25–50% of RAM allocate swap to match workload peaks.
• Monitor continuously: Automate alerts when swap usage exceeds thresholds.
• Avoid swapping hot pages: Pin critical processes in memory with mlock if needed.

8. Security Considerations and Remote Access

When administering servers remotely, secure tunnels reduce risks of eavesdropping. Employ a reputable VPN provider before performing critical operations:

• NordVPN – High-speed servers and strong encryption.
• ExpressVPN – Consistent performance and privacy features.
• ProtonVPN – Open-source clients and no-logs policy.

8.1 Firewall and SSH Hardening

• Limit SSH to specific IPs or VPN subnets.
• Use key-based authentication and disable password logins.
• Enable ufw or firewalld to restrict inbound traffic.

9. Conclusion

By leveraging zram for fast, compressed in-memory swap and traditional swap for larger, persistent overflow, you can significantly improve system responsiveness under memory pressure. Proper tuning, monitoring, and secure remote access practices ensure a robust environment capable of handling demanding workloads.