Requests Per Second
Learning Focus
By the end of this lesson you will know how to measure and interpret requests-per-second metrics.
What RPS Tells You
Requests per second (RPS) measures how many HTTP requests the server is handling. It is a primary indicator of server load and capacity.
Measuring RPS
From WebAdmin Dashboard
The dashboard shows real-time RPS on the main status page.
From Logs
# Count requests in the last 60 seconds
grep "$(date +'%d/%b/%Y:%H:%M')" /usr/local/lsws/logs/access.log | wc -l
# Average RPS over the last minute (divide by 60)
Interpreting RPS
| RPS Range | Server Status |
|---|---|
| 0-10 | Low traffic or idle |
| 10-100 | Normal for small-medium sites |
| 100-1000 | Active site, good throughput |
| 1000+ | High traffic, monitor resources carefully |
info
Cached requests are much cheaper than dynamic requests. A server can handle thousands of cached RPS but only tens of dynamic PHP RPS.
Key Takeaways
- RPS is a primary throughput indicator — monitor during peak hours.
- Distinguish between cached and dynamic RPS for meaningful analysis.
- Sudden drops may indicate upstream issues; sudden spikes may indicate attacks.
What's Next
- Continue to Connection Usage for connection-level metrics.