Will Microsoft Fabric replace Azure Synapse Analytics?

Overview of Microsoft Fabric

In the early months of 2023, Microsoft introduced Fabric in Public Preview. On 15^th Nov 2023, they announced the General Availability of Microsoft Fabric. For data professionals already using Synapse as a Platform-as-a-Service (PaaS) version, here is a small effort to explain what Fabric means and what its future is.

Microsoft Fabric is a Cloud-centric, all-in-one big data analytics service that allows us to design, implement, and maintain systems that collect, store, process, and analyze large volumes of data. Fabric offers a comprehensive suite of services through experiences such as Synapse Data Engineering, Synapse Data Science, Synapse Data Warehouse, Synapse real-time analytics, Data Factory, Data Activator, and Power BI.

For Fabric, data, and workloads are stored in the centralized repository, OneLake- a single, unified, logical data lake for the whole organization. OneLake allows you to virtualize your existing data lakes within Fabric. All fabric engines come with separate storage from computing and store their data in OneLake using a single, open data format, i.e., delta lake format. Fabric comes with the flexibility of either Lakehouse or data warehouse architecture implementation or a combination of both to get the best of both with simple implementation.

Things to know about Fabric

• To use MS Fabric service, you need a Power BI account and to sign up for the free Microsoft Fabric trial. If you don’t have a Power BI license, sign up for a Power BI free license and then use the Fabric trial.
• Lake-centric and open: The following picture shows the end-to-end architecture of the Lakehouse with components.

• Unified analytics platform: Here are components of MS Fabric,

1. Data Engineering: Fabric’s big data analytics workload allows us to leverage the power of Apache Spark for large-scale data transformation and build out a Lakehouse architecture.
2. Data Factory: It combines Power Query and Azure Data Factory into a data integration experience to perform the ETL (extract, load, and transform data) process. Powerful data orchestration allows us to build simple to complex data workflows with 300+ native connectors for integration.
3. Data Science: Empowers data scientists to explore, build, and manage data-learning models with integrated Azure machine learning.
4. Data Warehouse: Next generation of data warehousing in Microsoft Fabric with Delta Lake columnar storage engine for data with a SQL endpoint, fully separating compute from storage.
5. Real-Time Analytics: Provides authentic streaming experience in Fabric. Builds on Azure Synapse Data Explorer Synapse Real-time Analytics to implement streamlined data analysis.
6. Power BI: Provides Data Analytics and Data Visualization experience.
7. Data Activator: No code platform to identify patterns and trigger actions (Teams messages, emails, Power Automate tasks, etc).

• Fabric is AI-powered: Microsoft announced the public preview of Copilot in Fabric, which can be used with the Power BI, Data Factory, Data Engineering, and Data Science experiences. With Copilot in Fabric, we can create dataflows and pipelines, write SQL statements, build reports, or even develop machine learning models using natural language.

The Fabric Vs. Azure Synapse Analytics

Conclusion

Currently, Microsoft has no plans to retire Azure Synapse Analytics. Microsoft will provide advanced notice and adhere to the commitments to support the Modern Lifecycle Policy. Also, for data professionals, there is no straightforward best choice between Azure Synapse and Microsoft Fabric, as both can be used for different business needs. Organizations need no immediate change, and they can continue to use Microsoft’s current PaaS offering, i.e., Azure Synapse Analytics, to deploy, operate, and expand it. However, if an organization starts thinking of Microsoft Fabric as a future for data analytics solutions, then it is important to go with a thought process that considers factors such as data processing requirements, architecture, scalability requirements, and existing tools.

About CloudThat