Question

IH Purge

Is there a pre-defined methodology to be followed while purge IH records?

What sort of a performance impact would a growing IH cause while using an "Import IH" shape in strategies?

Thanks in advance,
Ratan Balaji.

Comments

Keep up to date on this post and subscribe to comments

September 11, 2019 - 7:54am

Hi Ratan,

As you might suspect, performance can be severely hampered by the record count. I would start by looking at the sample scripts Pega provides. 

https://community.pega.com/sites/default/files/help_v81/procomhelpmain.htm#dsm/tasks/dsm-ih-using-sample-scripts-tsk.htm

You're also going to need to consider extension table if applicable as well as adjusting your "Actuals Window" for ADM - if used.

Using KPI data is helpful in seeing when this kind of maintenance is needed. https://community.pega.com/sites/default/files/help_v81/procomhelpmain.htm#dsm/references/dsm-vbd-kpi-ref.htm

 

September 12, 2019 - 8:32am

Hi Paul,

Although the information is a good start for me, i would like to understand the reason behind such a process of purging.

September 12, 2019 - 9:26am

In real time applications, IH will grow in huge numbers. Example in my application it has about 180+ million in just 6 months in Production. So leaving it long would drastically effect performance. So archiving/purging this table is very important and you can work with your database admins to set up jobs on the database to do this activity.

Pega
September 12, 2019 - 10:26am

Purging IH on a periodic basis based on your business requirement keeps the performance optimal(execution of strategies, dataflows etc).
As the Interaction History grows very fast, if you don't control it then it may lead to processing huge amount of data which may impact the performance.

It is better to use the scripts shared above but if you chose to delete manually consider below points

Deleting records from the pr_data_ih_fact table won't have any impact on the application.
If you delete any record from the dimension table make sure to delete corresponding records from the fact table that references this data.

 

September 19, 2019 - 6:11am
Response to nawas

Hi Nawas,

Thanks for the response. What are the potential areas of impact due to a "growing IH"? Can you quote few examples?

Pega
September 13, 2019 - 3:13am

Hi Ratan,

The reason behind the purging of Pr_Data_IH_Fact is that interaction history tables contain transactional data which may grow fast. 

Therefore this table size increases very fast as a result sql query takes time to execute on this table sometimes it causes slowness or performance degrade of the application. So data older than one month should be purged from this table.

Thanks,
Abhinav

 

 

 

September 19, 2019 - 6:06am

When i asked to understand the process of purging, what i actually meant is NOT "why purge the data?" but "why purge it using such scripts?"