cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Analysis response time purepath

rguerrero
Inactive

Hello everyone;

I am analyzing a pp and is showing the following.


Maybe this is because I am monitoring only part of the entire infrastructure as shown in the image "0.jpg".

But the same should show me the total of the transaction in the transaction flow.

It shows me in the IIS layer 2.83 sec, then in the F5 layer 2.82 sec and in the BD layer 1.89 seconds and the transaction complete only lasts 2.83348 min. 2.jpg

What do you think the problem would be?

Thank for your help.

4 REPLIES 4

Joe_Hoffman
Dynatrace Champion
Dynatrace Champion

This can happen when some of the PP is making Asynchronous calls. Take a closer look at the call tree and you’ll likely see async calls.

rguerrero
Inactive

Hello Joseph;

this is happening for all purepaths, also the elapsed time shows the difference between the time of the asynchronous processes.

Hi Roberto, Could you share the PurePath session? My primary opinion about performance is:

1. The Web server or any tier may run into timeout but not the underlying backend which will be displayed in TFlow view as transaction taking more time on other layer than web.

2. The purepath length is HUGE (1821) this means that transaction flows across too many systems that need to be checked for performance problems. And hence complex transaction flow.

3. I believe, many of the redundant flows can be eliminated by appropriate architectural changes such as the TIER - SPLA-WSTSO5 -- Why transaction has to jump withing same TIER more than one instance? Why the each instance of same Tier has to make DB calls separate from the making thired party calls?

Best Regards, Rajesh

Hello Rajesh;

Thanks for your answer.

I share the session.

Regarding the communication between internal processes, this is implemented at the architecture level of the application.

Thanks for you help.

RGuerrero

purepath.dts