Evaluation question

Code profiling for .NET Developers

Moderators: StephenC, Alex.Davies, AndrewH

Evaluation question

Postby fratting » Wed Dec 30, 2009 2:04 am

I noticed looking at some of the times to execute a method that if method A calls into method B I may see method A took .340 seconds to execute but when I drill into method B I see a point where it may have taken .530 seconds to take. The point being method B took longer than method A, which doesn't seem to make any sense. Shouldn't the total time to execute method A be at least or larger than the time to execute method B? I'm looking at total time and hit count was 1 (I'm pretty sure). Is there a case when I may expect this to happen? Thanks.
fratting
 
Posts: 3
Joined: Tue Dec 29, 2009 7:01 pm

Postby Anu Deshpande » Thu Dec 31, 2009 4:27 pm

Many thanks for your post.

Can you kindly let us know which time column are you refering to is it Time (ms) or Time with Children (ms)?

Above information will help us to investigate further.
Anuradha Deshpande
Product Support
Redgate Software Ltd.
E-mail: support@red-gate.com
Anu Deshpande
 
Posts: 692
Joined: Mon Apr 20, 2009 3:53 pm
Location: Cambridge

Postby fratting » Wed Jan 06, 2010 6:59 pm

Hi,
I'm looking at Time(s)

Thanks,
Chris
fratting
 
Posts: 3
Joined: Tue Dec 29, 2009 7:01 pm

Re:

Postby Chris.Allen » Tue Jan 12, 2010 8:22 pm

fratting wrote:Hi,
I'm looking at Time(s)

Thanks,
Chris


Thanks - that explains it. The time for each method is the time spent exclusing in that method hence all values are cmpletely independent. "Time with children" however- should always decrease as you go down the call stack as this is cumulative time.

Hope that clears it up.
Chris.Allen
 
Posts: 594
Joined: Thu Mar 12, 2009 4:17 pm


Return to ANTS Performance Profiler 5

Who is online

Users browsing this forum: No registered users and 1 guest