Identity Fusion NG (tag: latest) Connector: Unable to handle large set of data

Hey Everyone.

What I am trying to achieve:

I want to aggregate all users from my baseline source (which consists of just over 450k accounts), merge them into the Identity Fusion connector based on the matching criteria, and correlate them.

I understand the account count will be different due to the accounts being merged, although there are many accounts in the baseline source that are still uncorrelated.

What I have tried in the source configuration:

  • Enabling Same-aggregation matching
  • Enabling an aggregation batch size
  • Enabling ''Include identities in the scope?" with a search query to pull the accounts from the baseline source.
  • Attempted to enable and use ‘Do not aggregate / Aggregate Before processing’
  • Enabling ‘Delete accounts with no managed accounts left’
  • Enabling ‘Skip accounts with missing ID’ (There should be no user with this missing)

Context

After retrying the aggregation a couple of times, we would also see it run for 6-8 hours and then fail.

I appreciate any suggestions on this.

Thanks.

I’m sorry to say that I’ve had the same experience, and that with a data set around 150k. I think the approach in the connector is good, but it’s an attempt to make the ISC solution do something way outside of its intended use, and it’s not sufficiently resourced for it. ISC is great at identity governance, but Identity Fusion is trying to perform the role of master data management.

I would love to hear from anyone who is successfully using the Identity Fusion connector since we have a lot of customers who would be interested in it. Even those with smaller populations.

Respectfully,
Matt

Hey @MattUribe,

Thank you for your response to this.

Just out of curiosity, based on your previous implementation with around 150k. What alternative route did you look into outside of the Identity Fusion Connector?

We’re still investigating. There are a couple of vendors that offer matching solutions, and there are some prior clients who have built their own custom solution.