Demystifying Linux Load Averages

Why does Linux show 3 load average values?

And when is your server actually under load?

If you’ve run uptime, top, or w on a Linux box, you’ve probably seen something like this:

load average: 0.35, 0.48, 0.51

So what are these three numbers?

They represent the average number of runnable processes over:

• 1 minute

• 5 minutes

• 15 minutes

Not CPU usage. Not memory usage. Just how many processes want CPU time.

⸻

So… when is your server actually under load?

Let’s say your system has 4 CPU cores.

If the load average is:

• 4.00 → Fully utilized

• >4.00 → Tasks are waiting for CPU — you’re under load

• <4.00 → Room to spare

For example:

load average: 5.60, 5.30, 5.10

On a 4-core system, this means your server has been overloaded for the past 15 minutes — there’s consistently more work than CPUs to handle it.

⸻

TL;DR:

• The 3 values show short, mid, and long-term load trends.

• Compare them against your CPU count.

• If they’re higher than your core count, your system is CPU-bound.

Knowing this helps with capacity planning, scaling decisions, and debugging performance issues — especially in distributed systems where one slow node can bottleneck everything.

What’s your favorite tool or approach to monitor system load in production?

Be the first to read stories and insights that inspire—subscribe to the newsletter and never miss a moment that could change your journey. — https://www.rasikmhetre.com/subscribe