Ubuntu – How to view the load average over a shorter time period

This site does a good job of explaining it. Basically, load average is the amount of traffic to your CPU(s) over the past 1, 5, and 15 minutes. Generally you want this number to be below the number of CPU(s)/cores you have. 1.0 on a single core machine means it's using the CPU to it's maximum, and anything above that means things are getting queued.

The CPU line in your top output is the current usage broken down by process types.

How about parsing the output from mpstat from the sysstat package? You can get statistics for each processor and set the interval (below is for 2s interval).

Sample output:

$ mpstat -P ALL 2
Linux 3.5.0-19-generic (ubuntu)     11/30/2012  _x86_64_    (4 CPU)

08:34:28 AM  CPU    %usr   %nice    %sys %iowait    %irq   %soft  %steal  %guest   %idle
08:34:30 AM  all    1.01    0.00    1.26    0.00    0.00    0.13    0.00    0.00   97.61
08:34:30 AM    0    2.50    0.00    1.50    0.00    0.00    0.50    0.00    0.00   95.50
08:34:30 AM    1    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00  100.00
08:34:30 AM    2    1.01    0.00    3.02    0.00    0.00    0.00    0.00    0.00   95.98
08:34:30 AM    3    0.50    0.00    0.50    0.00    0.00    0.00    0.00    0.00   99.00

Alternatively, you could implement the algorithm yourself. Take a look here and here.

Another way to do it would be to monitor the difference in idle time vs other usage time in /proc/stat at defined time interval and calculate the load based on that. Here's more info on /proc/stat.