Can you please help me with these . Would be great if you can explain with some examples. 1. Why link to link relationship is not recommended in RDV? 2. In BDV bridge table, if we are storing only hash keys( not natural keys), then how in fact/dimension we are going to get natural keys?
You need to add a relation describing hub between employee and project which contains the data for which employee works on a specific project. In your case you would have sattelite attached to the link. This is not advised in data vault modelling
Not creating an intertwined monolithic mess of a data structure, and adding possible support for streaming ETL. However, probably using a Data Mesh is even better as it's much easier to develop and doesn't have the stupid hash keys.
Erwin. No additional tools are required to design the model. You can design your approach to populate all hubs, links, data in parallel using Airflow. Since all these tables have hash key in it, there is no dependency across them.
Data Vault is crap ! It creates a plethora of tables for literally nothing. It's a non sens to add tables that doesn't bring real business value.. What is the purpose of dividing the model into hubs, links and satellites in the raw vault if at the next layer (business vault) we join again tables to consolidate them ? For developpers, a data vault model brings more complexity because they will need to do a lot of joinning to create there business agregated tables at the gold layer.
This is EXACTLLY what I confused!! The data vault, this kind of model designing no really avoid the nessary join. Although it seems clear in the staging layer, but in the downstream layer, we still need to use join to derive relative columns..
Very good explanation with a simple example
Thank you for a very detailed explanation. My question is why we need hash keys? Couldn't we just use the Business keys instead?
Great content ! How you migrated 3NF model into Data Vault model with example was amazing ! Why did you stop making further content ?
Remember that in real life settings you do not directly "convert" 3NF to DV or you end up with Source System Data Vault
A very good explanation, thank you so much!
Wow.. so nicely explained
Really well done please create more content
Absolutely brilliant explanation. Thank you very much
Can you please help me with these . Would be great if you can explain with some examples.
1. Why link to link relationship is not recommended in RDV?
2. In BDV bridge table, if we are storing only hash keys( not natural keys), then how in fact/dimension we are going to get natural keys?
Thanks. This gives better understanding at high level DV
Thanks, quite a nice explanation
Great introduction to Data Vault modelling
Thank you...This solidified my understanding
Thank you!
Nice, straightforward explanation. Thank you
You need to add a relation describing hub between employee and project which contains the data for which employee works on a specific project. In your case you would have sattelite attached to the link. This is not advised in data vault modelling
Excellent video !! Thanks.
Good one
good explanation
Well Done
Thank you. Very helpful.
bu aksan cok tanidik geliyor :)
where can I create that type of diagram of data vault ? I mean I need website name.
This is a great video. Thanks.
thanks!!
Very useFull, thank you
What is the special benefit of Data Vault modeling compared to convention EDW design say Kimball methodology ?
Not creating an intertwined monolithic mess of a data structure, and adding possible support for streaming ETL. However, probably using a Data Mesh is even better as it's much easier to develop and doesn't have the stupid hash keys.
what are the tools available in market to design Data Vault modeling
Erwin. No additional tools are required to design the model. You can design your approach to populate all hubs, links, data in parallel using Airflow. Since all these tables have hash key in it, there is no dependency across them.
Erwin
we use Enterprise architect at out organisation
Very nice!
Thank you
Turk musun? :)
Data Vault is crap ! It creates a plethora of tables for literally nothing.
It's a non sens to add tables that doesn't bring real business value..
What is the purpose of dividing the model into hubs, links and satellites in the raw vault if at the next layer (business vault) we join again tables to consolidate them ?
For developpers, a data vault model brings more complexity because they will need to do a lot of joinning to create there business agregated tables at the gold layer.
This is EXACTLLY what I confused!! The data vault, this kind of model designing no really avoid the nessary join. Although it seems clear in the staging layer, but in the downstream layer, we still need to use join to derive relative columns..