Breakdown of processing overhead for Dynatrace code module

[akataoka]

• Question
  I would like a breakdown of processing overhead for Dynatrace code module.
• Background
  When Dynatrace was introduced in the Openshift environment, the CPU processing time of the business application increased by about 30% compared to before the introduction of Dynatrace.
  We bileve that the above is due to the processing overhead of the Dynatrace code module injected into each pod.
  We would like to know the breakdown of processing overhead for Dynatrace code module, but there is nothing in the documantation, so we are having trouble explanning it to our customer.
  We need to explain the breakdown of processing overhead for Dynatrace code module in order to get customers to officially adopt Dynatrace.

[radek_jasinski · Accepted solution]

Hi @akataoka 

The impact on the performance of the monitored components should be hardly noticeable and should certainly not reach 30%. Only during the first start-up there may be a higher load as instrumentation, class browsing, etc. are performed.
There may be a number of variables that are causing the higher load and I recommend contacting support on this issue.

Radek

[yokada]

Hi, folks

Regarding of process CPU overhead of code module , we have been asked to explain from customer again and again. However we could not find related document and blogs in anywhere. In our experience,  process CPU overhead of Java code module seems be about 5 to 8 %.  Please share  actual data you have if possible.

 Regards