Linux — Delving into the Depths of Load Average

Whenever we encounter system sluggishness, our initial response typically involves running commands such as top or uptime to gauge the system’s load. For instance:

$ uptime
 10:36:07 up 20 days,  6:17,  3 users,  load average: 12.58, 10.94, 9.72

However, what I’m curious about is whether you understand the significance of each column in this output. You’re probably acquainted with the first three columns, denoting the current time, system uptime, and the count of logged-in users. But what about the last three numbers?

While you may recognize that the last three numbers represent the Load Average over the past 1, 5, and 15 minutes, have you ever delved into how these figures are computed?

I guess some of you may say that load average is the CPU usage per unit time, for example, load average: 0.63, 0.83, 0.88 means that the CPU usage is 63% for the past 1 minute. In fact, this is not the case.

Simply put, the load average refers to the average number of processes that in a running and an uninterruptible state per unit time, that is, the average number of active processes, which is not directly related to the CPU usage. Let’s explore more.

The Importance of Understanding Load Average