Skip to main content
Guidelines for scaling platform components based on load and resource requirements.

​
Vertical scaling

Increase system resource limits and tune configurations to handle more load on existing servers:
  • Raise file descriptor limits
  • Tune kernel network queues (somaxconn, netdev_max_backlog)
  • Increase worker processes and thread pools where supported

​
Configure file descriptor limits

  1. Edit /etc/security/limits.conf and add:
* soft nofile 500000
* hard nofile 500000
root soft nofile 500000
root hard nofile 500000
  1. Configure systemd defaults:
echo "DefaultLimitNOFILE=500000" | sudo tee -a /etc/systemd/system.conf
echo "DefaultLimitNOFILE=500000" | sudo tee -a /etc/systemd/user.conf
  1. Reboot to apply changes:
sudo reboot
  1. Verify:
ulimit -n

​
When to migrate to Kubernetes

Consider Kubernetes when:
  • MAU exceeds ~200k
  • You need multi-region deployments or failover
  • Sub-50 ms latency targets are critical
  • Dynamic autoscaling and elasticity are operational priorities (HPA/VPA)

​
Horizontal scaling guidelines

  • WebSocket Gateway: add ~1 replica per 1,000-1,500 peak concurrent connections (PCC)
  • Chat API: scale out when average CPU utilization exceeds ~60%
  • Kafka: increase partition count to improve throughput and parallelism
  • Redis: enable Redis Cluster mode when deployments exceed ~200k MAU to distribute data and improve scalability