I found in the manual short explanations for the abbreviations. But I have no idea about what their values stand for (how large is good, how large is bad).

How do I know if the IO is too busy and what makes the load too heavy?


snapshot of top

closed as too broad by Thomas Dickey, Stephen Harris, Jeff Schaller, Anthon, G-Man Aug 25 '16 at 4:35

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.


There is no fixed value because it depends on the computer's specifications as well as the programs that are running on it.

What your snap shot tells us is that your machine is mostly loaded on CPU processing and not I/O bound.

I personally prefer the regular top to get a global view of the machine. Its wa (wait %) indicator in the header is a good way to see how I/O bound the machine really is.

A production server's load should normally be kept under 2 * number_of_cores to ensure it remains reasonably responsive. Obviously, most tasks will be processed faster if the load is below the number of cores.

  • @JuliePeletier hi, I attached the snapshot of top but didn't find wa in the header. Concerning 2 * number_of_core do you refer to the load averages of CPU? – Lee Aug 25 '16 at 0:35
  • wa is just below stopped and is at 0.0%. This machine is not loaded at all with 19% idle and no I/O wait. I'm even wondering why you'd worry about performance in such a situation. – Julie Pelletier Aug 25 '16 at 3:53
  • I saw wa, on the row of CPU, but it's related to disk IO? The pictures are just for demonstration purpose. – Lee Aug 25 '16 at 4:44
  • Then I agree that this question becomes off-topic as unreproduceable and too broad. wa refers to I/O wait by the way. – Julie Pelletier Aug 25 '16 at 5:00

Not the answer you're looking for? Browse other questions tagged or ask your own question.