# I am trying to use debezium in a setup where each table is its own index. I want to create a merged index **URL:** \ **Category:** Elasticsearch\ **Created:** [November 18, 2024, 2:37pm UTC](https://discuss.elastic.co/t/i-am-trying-to-use-debezium-in-a-setup-where-each-table-is-its-own-index-i-want-to-create-a-merged-index/370700 "2024-11-18T14:37:05Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![randomuser24](https://avatars.discourse-cdn.com/v4/letter/r/919ad9/32.png) [@randomuser24](https://discuss.elastic.co/u/randomuser24)\ **Post date:** [November 18, 2024, 2:37pm UTC](https://discuss.elastic.co/t/i-am-trying-to-use-debezium-in-a-setup-where-each-table-is-its-own-index-i-want-to-create-a-merged-index/370700/1 "2024-11-18T14:37:05Z")
In the debezium setup i'll have an index for each table. Lets say I have the three sources indicies an Owner index, a pets index, treats index that each pet eats. How can I have a merged index where it will include a owner and the pets they own, the treats the pets eat. I understand we can use the enrichment policies but is there a way to do a merged index. The messages will only hit the three source indexes. Example ```auto Owner index [ { "_index": "owner", "_type": "_doc", "_id": "1", "_score": 1.0, "_source": { "owner_id": 1, "first_nm": "John", "last_nm": "Doe" } }, { "_index": "owner", "_type": "_doc", "_id": "2", "_score": 1.0, "_source": { "owner_id": 2, "first_nm": "Jane", "last_nm": "Doe" } } ] Pet index [ { "_index": "pets", "_type": "_doc", "_id": "1", "_score": 1.0, "_source": { "owner_id": 1, "pet_nm": "Bolt", "pet_id": 1 } }, { "_index": "pets", "_type": "_doc", "_id": "2", "_score": 1.0, "_source": { "owner_id": 1, "pet_nm": "Sparky", "pet_id": 2 } }, { "_index": "pets", "_type": "_doc", "_id": "3", "_score": 1.0, "_source": { "owner_id": 2, "pet_nm": "Yellow", "pet_id": 3 } }, { "_index": "pets", "_type": "_doc", "_id": "4", "_score": 1.0, "_source": { "owner_id": 2, "pet_nm": "Winnie", "pet_id": 4 } } ] Treats [ { "_index": "treats", "_type": "_doc", "_id": "1", "_score": 1.0, "_source": { "treat_id": 1, "food_nm": "decent food", "pet_id": 1 } }, { "_index": "treats", "_type": "_doc", "_id": "2", "_score": 1.0, "_source": { "treat_id": 2, "food_nm": "alright food", "pet_id": 1 } }, { "_index": "treats", "_type": "_doc", "_id": "3", "_score": 1.0, "_source": { "treat_id": 3, "treat_nm": "best food", "pet_id": 2 } }, { "_index": "treats", "_type": "_doc", "_id": "4", "_score": 1.0, "_source": { "treat_id": 4, "food_nm": "ok food", "pet_id": 2 } } ] Merged index needed [ { "_index": "owner", "_type": "_doc", "_id": "1", "_score": 1.0, "_source": { "owner_id": 1, "first_nm": "John", "last_nm": "Doe", "pets": [ { "pet_id": 1, "pet_nm": "Sparky" }, { "pet_id": 2, "pet_nm": "Sparky" } ], "treat": [ { "pet_id": 1, "treat_id": 1, "food_nm": "decent_food" }, { "pet_id": 2, "treat_id": 2, "food_nm": "alright food" }, ] } } etc.. ] ``` ---
**Author:** ![Christian\_Dahlqvist](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/christian_dahlqvist/32/4617_2.png) [@Christian\_Dahlqvist](https://discuss.elastic.co/u/Christian_Dahlqvist)\ **Post date:** [November 18, 2024, 3:14pm UTC](https://discuss.elastic.co/t/i-am-trying-to-use-debezium-in-a-setup-where-each-table-is-its-own-index-i-want-to-create-a-merged-index/370700/2 "2024-11-18T15:14:18Z")
This seems like a question for the Debezium community as there is no way to merge indices the way you describe (possibly apart from using enrichment policies) within Elasticsearch. ---
**Author:** ![randomuser24](https://avatars.discourse-cdn.com/v4/letter/r/919ad9/32.png) [@randomuser24](https://discuss.elastic.co/u/randomuser24)\ **Post date:** [November 18, 2024, 3:19pm UTC](https://discuss.elastic.co/t/i-am-trying-to-use-debezium-in-a-setup-where-each-table-is-its-own-index-i-want-to-create-a-merged-index/370700/3 "2024-11-18T15:19:25Z")
Will do on messaging the debezium community. Is there a possibility to use the transform api to get something like I mentioned in the topic above. I just made an edit. ---
**Author:** ![leandrojmp](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/leandrojmp/32/107231_2.png) [@leandrojmp](https://discuss.elastic.co/u/leandrojmp)\ **Post date:** [November 18, 2024, 3:34pm UTC](https://discuss.elastic.co/t/i-am-trying-to-use-debezium-in-a-setup-where-each-table-is-its-own-index-i-want-to-create-a-merged-index/370700/4 "2024-11-18T15:34:18Z")
> [@randomuser24](#): > > Is there a possibility to use the transform api to get something like I mentioned in the topic above. No, you want something like a JOIN, which does not exist in Elasticsearch. You need to normalize your data before indexing it. ---
**Author:** ![Christian\_Dahlqvist](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/christian_dahlqvist/32/4617_2.png) [@Christian\_Dahlqvist](https://discuss.elastic.co/u/Christian_Dahlqvist)\ **Post date:** [November 18, 2024, 4:02pm UTC](https://discuss.elastic.co/t/i-am-trying-to-use-debezium-in-a-setup-where-each-table-is-its-own-index-i-want-to-create-a-merged-index/370700/5 "2024-11-18T16:02:51Z")
It sounds like you want to replicate a join or materialized view of some sorts. This is in my opinion something that should be done before data is ingested into Elasticsearch.