Existing universal database models plus#

Snowflake is an ANSI SQL RDBMS with consumption-based pricing, and supports tables and views like all the relational solutions on the market today. Snowflake features to use in a Data Vault Here is a simple example of what a Data Vault 2.0 model looks like:

Existing universal database models plus#

Satellites give you the adaptability to record history at any interval you want, plus unquestionable auditability and traceability to your source systems. Links are designed to be many-to-many structures that give you the flexibility to absorb any cardinality and business rule changes without re-engineering (and therefore without reloading any data). We call this technique “ passive integration.” For an enterprise that will have multiple source systems, these business objects and their descriptive data are easily tracked through semantic integration by business keys. Now, source systems can be thought of as the automation engines of business processes, which in turn track a business object’s relationships and data state (e.g., on, off, active, inactive, balance, etc.).

Satellites: Descriptive data for hubs and links (Type 2 with history)īecause hubs are repositories of business object identifiers (the business key), this makes hubs business-driven because business objects themselves are the focus of business capabilities.

Links: Unique list of associations/transactions that represent the unit of work of a business process.

Hubs: Unique list of business keys that represent business objects.

It was built to be a granular, non-volatile, auditable, historical repository of enterprise data.Īt its core is a repeatable modeling technique that consists of just three main types of tables: The main point here is that DV was developed specifically to address agility, flexibility, and scalability issues found in the other mainstream data modeling approaches used in the data warehousing space. It is a data model that is architected specifically to meet the needs of today’s enterprise data warehouses. The design is flexible, scalable, consistent, and adaptable to the needs of the enterprise. It is a hybrid approach encompassing the best of breed between 3rd normal form (3NF) and star schema. The DV is a detail-oriented, historical tracking and uniquely linked set of normalized tables that support one or more functional areas of business. This is the formal definition, according to the inventor Dan Linstedt: In fact, we support any type of relational design, including Data Vault.” What is Data Vault modeling?įor those of you who have not yet heard of the Data Vault System of Business Intelligence or simply Data Vault (DV) modeling, it provides (among other things) a method and approach to modeling your enterprise data warehouse (EDW) to be agile, flexible, and scalable. Me: “Well, yes, I can understand your confusion in that case, but the name has nothing to do with data warehouse design really. So the conversation often goes like this:Ĭustomer: “Can you do Data Vault on Snowflake?”Ĭustomer: “Well, your name is Snowflake so we thought that might mean you only support Snowflake-type schemas.” Turns out we have a few customers who have existing data warehouses built using a particular approach known as the Data Vault modeling approach, and they have decided to move into Snowflake. Well, the cool thing is that Snowflake supports multiple data modeling approaches equally.

Since I joined Snowflake, I have been asked multiple times what data warehouse modeling approach Snowflake best supports.

It has been updated to reflect currently available products, features, and functionality.

PLEASE NOTE: This post was originally published in 2019.