# Effective satellite for link of json source

**URL:** <https://forum.data-community.org/t/effective-satellite-for-link-of-json-source/1798>\
**Category:** Data Vault 2.0\
**Tags:** link-satellite\
**Created:** [April 4, 2025, 9:52am UTC](https://forum.data-community.org/t/effective-satellite-for-link-of-json-source/1798 "2025-04-04T09:52:12Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Rameshkumar](https://avatars.discourse-cdn.com/v4/letter/r/e9c0ed/32.png) [@Rameshkumar](https://forum.data-community.org/u/Rameshkumar)\
**Post date:** [April 4, 2025, 9:52am UTC](https://forum.data-community.org/t/effective-satellite-for-link-of-json-source/1798/1 "2025-04-04T09:52:12Z")

</div>

Hello everyone,

Need some suggestion on below scenario,  
We have built a LNK\_\_SUPPLIER\_\_COMPANY from a source table in which we have json columns. The company details are present inside one of a json column. So we have flatten the column and pulled out the company id from json column.  
Here for one supplier id, we have multiple company\_id’s and load date is same for all these records as it is flattened from single row from source.  
example:

 ![image](https://europe1.discourse-cdn.com/flex013/uploads/datavaultusergroup/original/1X/5b44d4671f62e60826a764c0a20c64fefa902557.png)

LNK don’t have any issue as SUPPLIER, COMPANY combination for hash key.

We try to build Effectivity satellite for LNK\_\_SUPPLIER\_\_COMPANY, where we had driving key as supplier\_id and we ended up in primary key error as we have two company\_id, same load\_date for one supplier\_id.

Do we have any option like multi active effective satellite to handle this? any suggestion would be appreciated.

Note: Building effectivity satellite to capture latest updated company\_id for supplier\_id.

Thanks

---

<div class="post-metadata">

**Author:** ![patrickcuba](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.data-community.org/patrickcuba/32/19_2.png) [@patrickcuba](https://forum.data-community.org/u/patrickcuba)\
**Post date:** [April 7, 2025, 3:01am UTC](https://forum.data-community.org/t/effective-satellite-for-link-of-json-source/1798/2 "2025-04-07T03:01:04Z")

</div>

Hello,

I’m not sure how AutomateDV have done this, but an Effectivity Sat should handle 1:M relationships, scenario 3 in this blog

> **[The Link between Effectivity Satellites and Driver Keys.](https://medium.com/the-modern-scientist/the-link-between-effectivity-satellites-and-driver-keys-ad01ce2196e5)**
>
> Within the Data Vault community there is a misconception of what an effectivity satellite table is, how to model it, where it should be…

A link-sat with a finer grain than the link-hash key should then be a link-sat with a dependent-child key

> **[The Different Grains of Multi-Active Records.](https://medium.com/the-modern-scientist/the-different-grains-of-multi-active-records-832799677a8f)**
>
> Common feedback when data professionals learn of what a data vault model is, is that they foresee the large number of tables being…

As for the payload, if you can schematise before ingestion this will be cheaper and faster for processing semi-structured data, (from the screenshot this looks like Snowflake) [Schema detection and evolution for Kafka connector with Snowpipe Streaming | Snowflake Documentation](https://docs.snowflake.com/en/user-guide/data-load-snowpipe-streaming-kafka-schema-detection#configuring-required-kafka-properties)