Account Aggregation issue

Which IIQ version are you inquiring about?

Version 8.4

Share all details related to your problem, including any error messages you may have received.

Hello Experts,

We are working on Account Aggregation for Oracle Databases. we have 15 oracle databases and we have includes them into the Single Account Aggregation task.

We have analyzed that it is taking more than 7-8 hours to execute the task. Can you please help with suggestion that how to improve the performance. below are the details.

We have marked disable optimization as we want to aggregate whole target accounts.

Application Name Timing Total Records
Account Aggregation - Oracle DB (CAC) 4.17hrs 10524
Account Aggregation - Oracle DB (CBA) 2 mins 80
Account Aggregation - Oracle DB (CBR) 2 mins 107
Account Aggregation - Oracle DB (CO3) 6 mins 484
Account Aggregation - Oracle DB (CO4) 7 mins 561
Account Aggregation - Oracle DB (CO5) 5 mins 323
Account Aggregation - Oracle DB (CO7) 2.08 hrs 15419
Account Aggregation - Oracle DB (CWB) 5 mins 326
Account Aggregation - Oracle DB (DWH) 3 mins 216
Account Aggregation - Oracle DB (ETA) 2 mins 101
Account Aggregation - Oracle DB (ODS) 17 Mins 1745
Account Aggregation - Oracle Database (SCD) 14 Mins 789
Account Aggregation - Oracle Database (SSD) 10 Mins 585
Account Aggregation - Oracle Database (W08) 16 mins 1230
Account Aggregation - Oracle Database (W06) 24 mins 2127

We have enabled the partitioning.
Please help us to fix this issue on priority.

Optimization will still aggregate all of the accounts, what will happen is it will run the customization rule, the correlation rule, then compare all of the fields with the incoming aggregation object to the matched link object in IIQ to determine if any values have changed. If not, it will not submit the aggregation event for that account as it is not needed. This is going to be the most effective way to improve your performance.

It is also possible you have inefficient logic in your customization/correlation rules, in which case you can check this by using the Meter api to generate statistics that you can view on the debug page (call timings).

Hi @niket345,

Firsto of all, I suggest to unmark the disable optimization flag; if this flag is unmarked don’t update on SP the attributes that have not changed.

Later,
JDBC connetor launch the query on db and later processes the data.
Usually a complex query may be late 1-2 min, but it depends on its complexity.
In process data phase, only Sailpoint machines works.
So, you can check the execution time of those query with a db manager for exclude Oracle DB.
Later, you can check how much data are processes in SP, if you some transformation rule or somethings else.
Usually, the problem could be the quantity of data it must process and the resource of DB.
You can “fix” those times, removing the columns not necessary and increasing the resource(RAM & CPU) of the Sailpoint DB server.

After all, I suggest to you,make a deep analysis of the querys, resource usage and customization/creation rule.

As we dont have customization and correlation rule .we are using OOTB correlation config.

meterAPi is new to me. can you please provide steps to troubleshoot via meterAPi

We can not unmark the Disable optimization as we are using this to correlate reassignment and we are using OpenLDAP connector for Oracle Databases Application

Did you define maxThreads in RequestDefinition → Aggregate Partition ?
Also how many task server you have .
based on result you share i think even one aggregation is taking time , its doesn’t seems like problem with sequential task .

I would say 1st focus on any one of the account aggregation and see how you can optimize , then automatically sequential task will get optimized

Yes maxThread value is 1 under RequestDefinition - Aggregate partition. is this correct value configured?

We have 1 Task server, 1 UI server and 1 IQService in DEV
We have 2 Task Server, 2 UI Server and 1 IQService server in QA.

update that to 4/6 and now run the aggregation with partition for one account aggregation to see any improvement .

its still same…6 mins for 600 accounts. no improvement.

try for larger number of account (Account Aggregation - Oracle DB (CAC)) to know real effect . Also may be try increasing to 6 thread

How many partition its creating for 10524 accounts ?

What is your chunkToFetch value in the app definition? Also second thing did you try to install most recent OracleDriver?

@niket345 pls check with sailpoint support. here is similar type of requirement on below thread.

Aggregate data from multiple databases for certification - Compass (sailpoint.com)

Hi @niket345,
As you mentioned this is in non-prod environment.
May be in your prod environment you have more task servers, with partitioning enabled and maxThread allocations to aggregation partition will ideally should speed the aggregation timings.
If its not decreasing, need to enable loggers to see why is it taking time, are there any network issues with this oracle database from your sailpoint environment.

@vishal_kejriwal1 it is using default partition value as 1000 objects and took 10 partition in aggregation with execution time is 2.5 hrs

@kjakubiak do you have any sample configuration for chunkToFetch then i can configure same in App definition.

also we are using ojdbc11 jar for oracle19. Please suggest me that is it suitable jar for this or need to upgrade.

Have you written any custom rule , which get trigger during aggregation ?

@vishal_kejriwal1 we dont have any customization rule.

can you paste application xml and aggregation xml here .

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.