In the previous posts in the series, we have looked at the core resources available in Azure for Storage, Compute and Networking. In this post, we will look at the database services available.

Database Services

Azure Cosmos DB

Azure CosmosDB is a globally distributed service that supports a multi-model database service. It is Microsoft’s fully managed NoSQL database offering, but the multi-model service means it offers several different popular APIs for your teams to run their queries against. These APIs include SQL, MongoDB, Cassandra, Tables and Gremlin.

The wide range of APIs reduces the risks during platform migration as your developers can continue to query against the most appropriate APIs. The managed distributed service means you can have greater control over your scaling and redundancy, ensuring the database files are stored in the regions you require with the correct database throughput to ensure high responsiveness to queries.

Azure SQL Database

Alongside the NoSQL database service Azure CosmosDB, Microsoft also provides a version of their popular relational database SQL Service. Azure SQL Database is a platform as a service (PaaS) database service, which allows you to run SQL Databases on a service where upgrading, patching, monitoring and backups are completed without any user involvement. Therefore, you can concentrate on ensure the business intelligence and workflows are correct within your database.

Azure SQL Database offers a highly available and performant infrastructure to host your databases on, and Microsoft offers services like Azure Database Migration Service, which allows the migration of databases with minimal downtime affecting your systems.

SQL Managed Instance

Another option for hosting a SQL database is SQL Managed Instances. Like Azure SQL Database, SQL Managed Instances is another PaaS service offered by Microsoft, but with some crucial differences. SQL Managed Instances give you greater control over the Database Engine configuration used to host your Databases.

Azure Database for MySQL

Alongside the hosted SQL server offerings, Microsoft also provides several other Database engines too. Azure Database for MySQL is an example of another relational database service. It is based on the MySQL Community Edition. Like Azure SQL Database, the hardware is managed by Microsoft directly, so you do not have a team decided to host administration. The platform is highly available, and several service tiers are offered to ensure the platform can scale to the needs of your application.

Azure Database for PostgreSQL

Another example of a relational database service hosted by Microsoft is Azure Database for PostgreSQL, which is based on the community version of the open-source PostgreSQL database engine.

Azure offers two deployment options Single Server or Hyperscale. The Single Server option provides a similar service to Azure Database for MySQL. This highly available and scalable service is managed by Microsoft, giving you time to focus on the application needs. The Hyperscale (Citus) option horizontally scales across multiple machines, introducing the concept of sharding to run queries across these servers for faster response times on larger datasets (greater than 100GB).

Big Data Services

Alongside the database platforms Microsoft provides, they also supply a range of services that can be used for Big Data analysis designed for larger datasets of information unsuitable for traditional forms of data processing.

Azure Synapse Analytics

Azure Synapse is an enterprise-level service that provides analytic services that can be run on your data warehouses. They offer both serverless and dedicated resource models, allowing you to access always available resources for burst workloads, or reserve resources for standard workloads, giving you predictable costs.

These services can be integrated with other Azure services like PowerBI or Azure Machine-Learning.

Azure HDInsight

Azure HDInsight is a cloud-hosted, managed, open-source analytics service that supports various open-source frameworks including, Hadoop, Spark and Kafka. These frameworks allow for batch processing (Extract, Transform and Load), data warehousing, Internet of Things (IoT), Data Science and Machine Learning scenarios.

Azure Databricks

Azure Databricks is a data analytics platform optimized for Azure and can help unlock insights of your data and build AI (Artificial Intelligence) solutions. It supports a range of data science frameworks, PyTorch, TensorFlow, and a number of languages Python, Scala, R, Java, and SQL.

Azure Data Lake Analytics

Azure Data Lake Analytics is an on-demand service to run analysis on big data. You write queries to process your data, while Microsoft provides the platform that can scale to the required level. As the service is on-demand, you only pay for the resources used while your queries are running.

Summary

In this post, we have looked at various data services that can be used to host, store and query your data. Whether your data is structured relational data, unstructured data, or the dataset is too large for traditional forms of processing to be practical.  

— Matt Boyd