👉Learn Module: Explore concepts of non-relational data
👉Non relational data has no proper structure in them.
👉No scheme on Data.
👉Store the information for entities in collections or containers and not in relational tables.
👉Two entities in the same collection can have a different set of fields.
👉Data retrieval capabilities of a non-relational database vary based on stored data.
👉Advanced non-relational systems such as Azure Cosmos DB support indexing.
👉Use Cases include IoT and telematics, Retail and marketing, Gaming and Web and mobile applications.
👉Two types of non-relational data
- Data contains fields and two entities of same type can have different fields.
- There are different formats of semi-structured data.
- JSON =Document enclosed in curly brackets.
- Avro =Row-based format.
- ORC = organizes data into columns rather than rows.
- Parquet = columnar data format.
- Do not have fields.
- Ex: Audio and Video
👉Non-relational database also known as No-SQL database.
👉There are four types of non-relational database.
- Each data item in a key-value store has two elements, a key and a value.
- Key uniquely identifies the item and valued hold the data.
- value is opaque(DBMS just sees the value as an unstructured block).
- Ex: Azure Table storage
- Each document has a unique and fields in the documents are transparent to the DBMS.
- Store data in JSON format or other formats such as XML, YAML, and BSON
- Ex: Azure Cosmos DB
3.Column family database
- Organizes data into rows and columns. Ex: ORC and Parquet files
- Conceptually similar to relational database but different in implementation.
- Ex: Apache Cassandra in Azure Cosmos DB supported through the Cassandra API
- Store entities with a main focus on the relationship between these entities.
- stores two types of information. Nodes = Instances of entities and Edges= Relationships between nodes.
- Nodes and edges can both have properties that provide information about that node or edge.