Kernel manages operating system resources. User program can only access to those resources by making system calls to the kernel. System call is similar to an API of kernel, which in term, runs kernel tasks your program needs.

str = "something" // run on user space
x = x + 1 // run on user space
file.write(str) // switch to kernel space
y = x + 4 // switch back to user space

Use top command to exam CPU time allocation between user space and kernel space.

Explanation:

1. 24.8 us (user space): 24.8% of CPU time is spent on user process.
2. 0.5 sy (system): 0.5% of CPU time is spent on kernel space.
3. ni (niceness): the ratio of CPU time spent on low priority processes.
4. id (idle): the ratio of CPU time spent on idle processes (during the period of time, CPU can not do anything).
5. wa (wait): the ratio of CPU time spent on waiting for I/O (during the period of time, CPU can not do anything).
6. hi (hardware interrupt): the ratio of CPU time spent on responding hardware interruption.
7. si (software interrupt): the ratio of CPU time spent on responding software interruption.
8. st (stole time): the ratio of CPU time stolen by virtual machine.

Use time command.

time ./perl-timeout-example 100.100.100

We could not ping the desired address!

real    0m5.0013s
user    0m0.004s
sys     0m0.008s

• Real: Total time from start to finish of the call, including the CPU time spent on other processes.
• User: The amount of CPU time spent in user space.
• Sys: The amount of CPU time spent in the kernel space.

Normally, user + sys is the total CPU time spent on such process. It is smaller than real. But in multicore CPU, user + sys is the total CPU time spent on this process. It could be large than real.