• State Verified Answer
  • Locked Locked
  • Replies 3 replies
  • Subscribers 58 subscribers
  • Views 4539 views
  • Users 0 members are here
Options
Related
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

PA vs PA

george.avramis
over 6 years ago

Hello all.

In PA client we could select to see the Raw data so that when selecting to view the last hour data we would see lots of spikes in the graphs signalling that there had been no aggregation in the data.

In PI when looking at any graph it is always smooth no matter if it is Raw or aggregated.

We have a customer that is asking if it is indeed raw data that we are seeing or if it has been aggregated no matter what we select.

Any ideas?

 

Thanks in advance.

Parents
  • +1 over 6 years ago
    Hi George,

    The sampling mechanism is different between PA and PI. The PI sampling mechanism works on a collection frequency mechanism and captures SQL statements and data from the database instance dictionary views. PA captures data using collectors installed directly on the host from the server memory.

    We are working on improvement of providing more detailed statistics, such as wait events, as a part of enhancement of PI abilities for the next release.

    I would recommend opening a Service Request with Support if you require more details about specific metrics and data.

    John
Reply
  • +1 over 6 years ago
    Hi George,

    The sampling mechanism is different between PA and PI. The PI sampling mechanism works on a collection frequency mechanism and captures SQL statements and data from the database instance dictionary views. PA captures data using collectors installed directly on the host from the server memory.

    We are working on improvement of providing more detailed statistics, such as wait events, as a part of enhancement of PI abilities for the next release.

    I would recommend opening a Service Request with Support if you require more details about specific metrics and data.

    John
Children
No Data