Summary of The Developer's Guide to Data Modelling with Document Databases - Adrienne Braganza Tacke

00:00:00 - 00:40:00

This video discusses the importance of data modeling with document databases, and provides a step-by-step guide on how to create a data model for a document database. It also covers the three phases of data modeling: inputting data, mapping entities and relationships, and applying design patterns.

• 00:00:00 The speaker discusses the importance of data modeling with document databases, explaining that although schema-less databases are flexible, this does not mean that you don't need a data model. The speaker provides a step-by-step guide on how to create a data model for a document database, using examples of a casino floor analytics application and Reddit posts.
• 00:05:00 The Developer's Guide to Data Modelling with Document Databases covers the steps of qualifying and quantifying operations, gathering logs and statistics, and mapping relationships. Once the data is understood, the next phase is to predict scenarios and create plans based on the information.
• 00:10:00 The video discusses how to model data using document databases: linking and embedding. Linking is similar to having a foreign key relationship, while embedding takes the information you need and embeds it into the parent document. If the type of relationship is one-to-one, one-to-many, or many-to-many, linking is the same across all of them; however, if you have a one-to-zillions relationship, embedding is the better option.
• 00:15:00 The video discusses the three phases of data modelling: inputting data, mapping entities and relationships, and applying design patterns. The third phase is an iterative approach, where if the data model that is created in phase two is working, then no design patterns are needed. If new data is needed, or the complexity of applying design patterns outweighs the simplicity of the data model, then phase three is needed.
• 00:20:00 The "Attribute Pattern" can be used to reduce the amount of data needed to be searched when querying documents, by only querying against fields that are required for the current policy. Another assumption is that the documents do not have too many revisions.
• 00:25:00 This video discusses the different data modelling patterns, including the attribute pattern, polymorphic pattern, and document versioning pattern. It provides a brief example of how to apply each pattern to a casino floor analytics application.
• 00:30:00 The data model for a slot machine includes the types of slot machines present, the name of the casino, the amount of money the player has won, the floor manager for the casino, and the players present.
• 00:35:00 The video discusses how to model data for an analytics application, with a focus on simplicity vs. performance. If simplicity is the goal, then the model should be lightweight and fit within the resources available. If performance is the goal, then the model should be robust and accommodate high traffic rates.
• 00:40:00 This video introduces the concepts of data modelling and document databases, and explains how to use these tools to improve performance. It recommends using simplicity and performance as guides when designing a data model, and advises tracking changes in the data model as performance improves.