Relational And Non-Relational Datastores

There are two types of datastores or data store formats used by companies to store their data namely Relational datastores and Non-Relational datastores.

In this blog, we are going to cover about Relational and Non-Relational datastores, Azure Services related to each datastore type, when and where to use the particular datastore type.

What Are Relational Datastores?

Relational datastores have been the most prevalent technology for decades, they are proven and widely implemented. The Relational datastores provide a store of relational databases with related tables. The tables have fixed schemas of rows and columns. SQL (Structured Query Language) is used to manage data present in relational datastores.

It supports a schema-on-write model, wherein the table structure is defined ahead of time, and all the read or write operations have to use the schema defined or set for the table.

Azure Services Based On Relational Datastores

Azure offers four types of Relational datastores as a service. They are:

• Azure SQL Database: It is a part of the Azure SQL Family. It is a scalable, relational database service built for the cloud and is based on Microsoft SQL Server Database Engine. It is available with three deployment options- Azure SQL Database server, Azure Virtual Network, and Azure SQL Database Serverless.
• Azure Database for MySQL: It is a managed relational database service based on an open-source MySQL server engine.
• Azure Database for MariaDB: It is a fully managed relational database in the cloud. MariaDB is a popular open-source database server. It handles critical workloads with dynamic scalability and predictable performance.
• Azure Database for PostgreSQL: It is a fully managed relational database based on an open-source Postgres database engine. It supports various development platforms like Java, C++, Python, etc. It is available with two deployment options- Single server deployment and Hyperscale (Citus) deployment option.

Issues With Relational Datastores

We know data comes in varied formats like documents, JSON files, images, etc, and sizes. It is not feasible to store every data in rows and columns or in relational database structure for an organization. To solve this problem we have a concept of non-relational datastores.

What Are Non-Relational Datastores?

Non-Relational datastores are databases that do not have the tabular schema of rows and columns like relational datastores. They provide high performance, easy to use, and scalable.

Non-relational data stores are also known as NoSQL data stores.

Types Of Non-Relational Datastores

There are four types of non-relational datastores or NoSQL datastores available.

1.) Document datastores: Here the data is stored in the form of documents and each document consists of named fields and data. The data can be stored in any format like JSON, XML, YAML, or even a plain text file.

• Each document has its own schema.
• Document data is semi-structured i.e, data types of each field are not strictly defined.
• Documents are retrieved by unique keys.

Use case: Content Management, Product Catalog, Inventory Management.

2.) Columnar datastores: Conceptually it is very similar to the relational model, the data is stored in rows and columns but the columns are divided into groups known as Column-families. The columns are logically related to each other. They are also known as Wide-Column Datastores.

• The columns can vary by individual rows.
• Scalable
• Update and Delete operations are rarely performed in this data store.

Use case: Web Analytics, Recommendations, Messaging, etc.

3.) Key-Value datastores: It is a large hash table where each data value is associated with a unique key. A hashing function is used to provide an event distribution of the hashed keys across the data storage.

4.) Graph datastores: In this type of data store, two types of information is stored- nodes and edges where nodes represent the entities and edges represent the relationship between these entities. It is suitable for those applications which require to execute queries that need to traverse a network of nodes and edges and analyze the relationship between different entities.

Azure Services Based On Non-Relational Datastores

There are four Azure Services based on non-relational datastore models. They are:

• Azure CosmosDB: It is a globally distributes, multi-model database service which manages data as a partitioned set of documents. It is a serverless service and scalable.
• Azure File Storage: It allows us to create files shares in the cloud, and access them from anywhere with an internet connection.
• Azure Blob Storage: This service enables you to store massive amounts of unstructured data, or blobs, in the cloud.
• Azure Table Storage: In this service, the data for an item is stored as a set of fields, and the item is identified by a unique key. It implements the key-value datastore model.

Considerations For Relational And Non-Relational Datastores

Many times there is a doubt as to which datastore or model to select for the project. So here are some key points which will help you in selecting the best model for your project.

Consider the Relational datastore when:

1. Workload volume is consistent and requires medium to large scale
2. Working with complex queries and reports
3. Users are more centralized
4. The application will be deployed to large, high-end hardware

These are some scenarios where we can consider to use relational datastores:

• Accounting, finance, and banking systems.
• Inventory management systems.
• Transaction management systems.

Consider the Non-Relational datastore when:

1. Workload volume is high and requires large scale
2. Data retrieval is simple and tends to be flat
3. Wide geographic distribution
4. The application will be deployed to commodity hardware, such as with public clouds

These are some scenarios where we can consider to use non-relational datastores:

• Mobile apps.
• Real-time analytics
• Content management
• Personalization
• IoT apps
• Database migration

Related/References

• Microsoft Azure Data Fundamentals [DP-900]: All You Need To Know
• Implementing An Azure Data Solution Exam [DP-200]: Everything You Need To Know
• Implementing an Azure Data Solution | DP-200 | Step By Step Activity Guides [Hands-On Labs]
• [DP-100] Microsoft Certified Azure Data Scientist Associate: Everything you must know
• Microsoft Certified Azure Data Scientist Associate | DP 100 | Step By Step Activity Guides (Hands-On Labs)
• [AI-900] Microsoft Certified Azure AI Fundamentals Course: Everything you must know
• Microsoft Azure AI Fundamentals [AI-900]: Step By Step Activity Guides (Hands-On Labs)

Next Task For You

You will know more about data concepts, why you should learn, Job opportunities, and what to study to clear [DP-900] Microsoft Azure Data Fundamentals Certification by registering for our FREE CLASS.

Click on the below image to join the FREE CLASS for Microsoft Azure Data Fundamentals [DP-900].

The post Relational And Non-Relational Datastores appeared first on Cloud Training Program.