Understanding Linux CPU Load

Updated: 11 October 2026 · Applies to: AlmaLinux 9, Rocky Linux 9, Ubuntu 24.04 and Debian 12

The load average is the average number of processes that are running or waiting to run, over the last 1, 5 and 15 minutes. Linux also counts processes waiting for disk or network storage, so a high load does not always mean the CPU is busy.

Check the load

uptime
 10:42:01 up 12 days,  3:10,  1 user,  load average: 1.20, 0.85, 0.60

The three numbers are the 1-, 5- and 15-minute averages. You also see them at the top of top and htop, and in /proc/loadavg.

Compare it with the number of CPU cores

nproc

A load equal to the number of cores means the CPUs are fully used; above that, processes have to wait. On a server with 4 cores:

• 2.0: about half the CPU capacity is in use.
• 4.0: the CPUs are fully used.
• 8.0: on average, 4 processes are waiting at any moment, and the server feels slow.

Short peaks in the 1-minute value are normal. A 15-minute value that stays above the number of cores needs attention.

Find the cause

• Run top and press P to sort by CPU or M to sort by memory; the busiest processes are at the top.
• A high wa (I/O wait) value in top with low CPU use points at the disk. Check it with iostat -x 1 (from the sysstat package) or iotop.
• Processes with state D in top are waiting for I/O. Many of them explain a high load while the CPU is mostly idle.
• Check memory with free -h. When a server runs out of memory and uses swap, processes wait for the disk and the load rises.

If the load stays high and you cannot find the cause, open a support ticket and include the output of uptime, free -h and top -b -n 1 | head -n 20.