The number of processors used totally change my code profiling

Question

0.00/5 (No votes)

See more:

Hello everyone!
I have one simple question: why the number of processors used when I compile my code with a profiling tool doesn't decrease the time spent in each function? For example, if I compile the first time with 1 processors, and then with four, I expect to see the same flat profile (like the same percentage and the same most expensive functions etc..) but just with the time spent in each function divided by four. But it's not the case... I have a totally different list of functions and the time is not coherent. Why??
PS: I use Gprof.
Thanks!

What I have tried:

I tried to use one then two then four processors and every time I have a totally different profiling.. I really don't know what I can do and what does that mean.

Posted 28-May-18 22:46pm

Member 13849020

Updated 28-May-18 23:08pm

Add a Solution

1 solution

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

Daniel Pfeffer · Accepted Answer · 2018-05-28T23:10:00

Solution 1

A function is typically executed in the context of a single thread. Having additional processors won't improve its speed.

Additional processors are useful when your program can be divided into multiple threads. In this case, the more processors, the more threads may be executed simultaneously (true multi-tasking, not time-slicing). This has limits (see Amdahl's Law), and problems (e.g synchronisation between threads).

Multi-tasking requires a different approach to programming than single-tasking, but properly implemented it can significantly speed up many programs.

Posted 28-May-18 23:10pm

Daniel Pfeffer

Comments

Member 13849020 29-May-18 5:13am

Ok thank you very much, I understood that the number of processors and the speed of my program aren't directly related, but why the most expensive functions with one processors aren't the same with 4?

CPallini 29-May-18 5:35am

5.

Daniel Pfeffer 29-May-18 7:52am

One possibility is that with more processors, less switching-in and switching-out of threads is necessary. This means that the internal data/instruction cache of each processor is not flushed so often, which in turn means that the functions in general execute more quickly.

Note that this is only a guess. It would be impossible to give you a better answer without a full analysis of the profiling data.

Member 13849020 29-May-18 8:36am

yes of course! Thank you for your time!