AWS Amplify with Amazon DynamoDB Single Table and Direct AWS Lambda Resolvers with Rick Houlihan

[@gerrykaplan2289]

Is there another video like this that shows how to define the schema for a single table dynamodb and the lambda resolver from the source code level in the Amplify project?

[@rodrigobraden8960]

Pro trick : watch series at Kaldrostream. I've been using them for watching loads of movies lately.

[@briggsasher5478]

@Rodrigo Braden Definitely, been watching on KaldroStream for since november myself :)

[@piercejonah6289]

@Rodrigo Braden yup, I have been using KaldroStream for since november myself :D

[@datoubi]

@Gerry Kaplan did you find a good resource?

[@vgribok]

FYI: The example in this video shows how to achieve DynamoDB single-table design, but does it in a way that does not illustrate its main advantage: the ability to fetch multiple "pre-joined" types of entities that share same partition key. The example still does sequence of fetch-the-parent-then-children, which is not very efficient, and could just as well involve multiple tables without notable performance difference.

[@alankarmisra]

They discuss that possibility at 37:15 saying that you could return a hierarchical data structure by modifying the lambda function response. If you have a single query to get pre-joined entities from DDB based on the same partition key, you would then shape it in your lambda function as an array or an object or a list of entities depending upon what the graphql interface expects. At least that's what I made of it.

[@vgribok]

​@@alankarmisra Yes, they mention it's possible, but the video would be much more on point, IMO, if it was shown how to use partition key (with optional sort key) to get hierarchical GQL data in a single DDB roundtrip.

[@mohitpalyekar6729]

can you please share the resources?

[@sauer.voussoir]

how did you set the type column that has (!) beside it, the one with the entity table names data?

[@mtgoss40]

Why does the title say AWS Amplify but the entire video is spent in the AWS console?

[@slikk66]

where is that visualizer at 12:52 ?

[@eric85077]

NoSQL Workbench for DynamoDB

[@yacinehannane2908]

wish nader got more time to speak to be honest

[@uNki23]

I see zero benefits compared to an http endpoint that would actually make use of single table design and query an item collection containing multiple different entities with one single query to DDB. If I don’t do that, I’ll make multiple requests to DDB to fetch my data. This is also why I’ll never understand the hype for GQL. I’ll most likely have multiple requests to a database vs a single request with a specific endpoint.