Export to Data Lake Alternatives: Fabric Link Vs. Azure Synapse Link

We previously explored Microsoft's deprecation of the "Export to Data Lake" feature for Dynamics 365 Finance and Operations (F&O), what it means, and your options for ensuring a smooth transition.

In this blog, we explain and compare the two options organizations should consider for replacing Export to Data Lake to help you make the best decision for your organization.

Why Did Microsoft Deprecate Export to Data Lake?

Microsoft decided to sunset the "Export to Data Lake" feature for 2 main reasons.

  1. To streamline its data integration offerings and encourage the adoption of more advanced and flexible solutions.
  2. To move away from legacy replication technology called change data capture (CDC).

Export to Data lake also delivers data in CSV format, which can result in poor performance, and read/write errors when used directly for reporting, and requires multiple transformations to optimize.

Microsoft has made significant advancements in their technology stack since the launch of Export to Data Lake, introducing technologies that cut down on storage and compute costs, as well as reduce the steps necessary to move and transform data. With the newer approaches, customers can land their data directly in the OneLake as delta tables, unlocking a wave of new functionality when paired with Power BI and other tools.

When Will Export to Data Lake Get Shut off?

Export to Data Lake is deprecated as of November 1, 2024. It will continue to function with limited support.  Microsoft is allowing a one-time extension through February 15, 2025.  You can submit the request via the Azure Synapse Link option in Power Apps. You must submit a request for each environment. Microsoft plans to officially shut off Export to Data Lake beginning March 25, 2025.

We are advising organizations using this legacy method to not delay their migration and fully understand the implications of each option and start your transition to alternative solutions as soon as possible.

Organizations have 2 main options to consider: Fabric Link or Azure Synapse Link. Synapse link requires a whole lot more manual configuration than Fabric link, but it does give you the benefit of not having to pay in perpetuity for data storage. More on each option below.

Option 1: Fabric Link

Microsoft Fabric Link is a feature that enables organizations to connect their data from Microsoft Dataverse directly to Microsoft Fabric. This integration allows users to access and analyze their Dynamics 365 and Power Apps data seamlessly within Microsoft Fabric’s environment, specifically utilizing OneLake, the built-in data lake.

If you are already utilizing Power BI for data consumption, employing a data warehouse, or using dataflows and notebooks for data transformation, the link to Fabric feature offers considerable benefits. This feature allows you to streamline your data integration architecture by eliminating the need for a separate storage account or Synapse services for Dataverse data. Instead of incurring costs for Azure resources such as storage and compute, you will pay for the additional Dataverse storage required. Compute charges, including near real-time data updates and management overhead, are integrated into the Dataverse storage costs.

Fabric link is the primary Export to Data Lake alternative that everybody's talking about. But Fabric link has some serious cost considerations that people are not always aware of.

The biggest consideration is when you use Fabric link, is that it stores the data in an intermediate form in the Dataverse environment at a very costly rate of $40/gig/mo. or $40,000 / TB / Mo. Yikes!  This cost is only incurred when you exceed the storage capacity allotment provided for with your licensing agreement.  Additionally, data storage is compressed so you cannot directly compare it to current storage requirements.

Data latency for Fabric link is about 1 hour, which may be an issue for some use cases.  Microsoft is working to reduce that latency, but timeframes and new latency metrics are not yet available.

There is also a 1000 table limit with Fabric Link.  All Dataverse tables are automatically added to Fabric link, so if you have Dynamics CE or CRM, or have a large Dataverse presence, you may hit the limit.  Microsoft is also working to remove this restriction.

Option 2: Azure Synapse Link

Microsoft Azure Synapse Link is the non-Fabric based direct replacement for Export to Data Lake.  It makes Dynamics 365 F&O and/or Dataverse data available within an Azure Synapse environment or storage container. This approach has 2 options:

  1. CSV
    1. This option has the lowest data latency and only has storage costs but requires the greatest transformation effort to read and apply to an analytics solution.
  2. Delta
    1. This option requires a Spark Pool and runs Microsoft managed notebooks to create delta tables in Synapse from the source system. This simplifies the logic and effort to make the source data ready for use in an analytics solution.

Both Synapse Link approaches eliminate the high cost of Dataverse storage, using Azure Storage instead. However, you will incur compute costs to transform the data for downstream use.

OmniData has developed processes to simplify and quickly implement both options.  If you need low data latency and have complex transformation processes, we have a framework for ingesting the CSV files efficiently.

If you choose the Delta approach, we can help to implement and integrate with existing solutions or even with Fabric using data virtualization shortcuts.

Resource: Microsoft’s Azure Synapse Transition Guide

What is the best Option to replace Export to Data Lake Feature?

Both provide enhanced performance, greater flexibility, and new features such as Delta Table functionality. The comparison table below helps illustrate the differences between the two options:

Fabric Link Vs. Azure Synapse Link – A Comparison

Feature omnidata azure logo Azure Synapse Link omnidata fabric logo Fabric Link
Data Refresh Every 15 min Every 60 min
Table sync Selected tables/ entities All tables selected by default
Best for Medium to Large Enterprises: Better suited for data engineers and analysts needing comprehensive analytics solutions. Ideal for users looking for straightforward integration without extensive data engineering.
Data Storage Data Lake Parquet data in Dataverse
Storage Cost (~$26/TB/Mo) Spark pools for Delta conversion ($626 - $4176 / mo.) (~$40,000/TB/Mo) When over capacity allotment and after data compression)
Analysis / Querying Synapse Serverless Consumes Fabric Capacity
Shortcut to Fabric? Yes -
Data Flow Supports complex data transformations and analytics. Directly streams data into OneLake for Power BI and Fabric-based reporting.
Integration Complexity More complex, suitable for high-frequency, high-volume data scenarios. ETL-free, designed for ease of use with Dynamics 365 data integration.
Performance Optimized for handling large datasets and complex queries efficiently. Focuses on seamless data integration with good performance for standard use cases.
Use Cases Suitable for advanced analytics, data warehousing, and large-scale data processing. Best for operational reporting and business intelligence.

How to Prepare for Your Export to Data Lake Migration

While a trusted Microsoft partner can handle the physical migration process, there are several important steps your organization needs to take to ensure a smooth transition from Export to Data Lake.

1. Determine Dependencies

It's crucial to evaluate your current data setup and figure out what depends on the Export to Data Lake feature.  Define clear objectives for the migration, like whether you want better performance, lower costs, or enhanced analytics.

2. Create an Internal Migration Plan

Next, develop a detailed migration plan internally that includes timelines, and configure necessary settings in your Dynamics 365 environment to avoid any disruptions. This is something we can assist with if you don’t have the bandwidth, but it is critical to align your team internally with a proper plan, including communications.

3. Allocate Resources

One of the most overlooked aspects of these migrations is allocating resources to the project. Ensuring that you have the right people and resources available at the right times is critical to prevent delays and ensure the project stays on track. By taking these preparatory steps, you'll be well-positioned for a successful migration and can fully leverage the benefits of the new system.

This migration can happen in just a couple of weeks. But it all depends on having a solid timeline and the right resources. You don’t want to end up stuck with no resources, causing the project to drag on for months. Make sure you tackle potential risks ahead of time, especially around planning and switching over to the new system. Have a clear plan for how that cutover will go down.

4. Prep Your Dynamics 365 Environment

Finally, make sure all necessary configurations are in place in your Dynamics 365 environment, such as enabling SQL Row Version Change Tracking.

Our Export to Data Lake Replacement Process

At OmniData, we pride ourselves on delivering exceptional value to our clients through a well-defined and efficient migration process. Our approach ensures that your transition is smooth, secure, and optimized for your specific needs.

We start by setting up either Azure Synapse Link or Fabric Link, depending on your choice. We ensure your environment is ready for integration and ensure it is aligned with your data governance framework to manage data quality and security.

We provide training for your team on the new tools and processes and document the transition steps. We conduct thorough testing to ensure data integrity and performance and validate that the new setup meets your analytical needs.

We'll make sure your reporting keeps running smoothly with no downtime when you switch to production. We'll help you plan everything out, so it's seamless and hassle-free. And we're here to help every step of the way.

After the transition, we continuously monitor the performance and adjust as necessary to optimize data processing and analytics.

Next Steps

By following these steps, you can facilitate a smoother transition and leverage the enhanced capabilities of either Microsoft Fabric Link or Azure Synapse Link.

We understand that choosing the right tools for your data needs can be challenging. Whether you are still unsure of the best path forward, or are ready to make the transition, our team is well-versed in the data ecosystem and can assist you in executing a seamless transition.

Whether you need help with data integration, optimizing performance, or setting up real-time analytics, we are here to support you every step of the way. Get in touch and let us help you achieve your data goals efficiently and effectively.

omnidata square logo

OmniData

OmniData is a leading Microsoft partner and professional services firm specializing in data and analytics modernization and AI solutions. We provide actionable insights and scalable solutions powered by AI, seamlessly integrating with existing systems to maximize efficiency. Our team of veteran specialists excels in solution architecture, data engineering, and business intelligence. We thrive on partnering with companies across various industries and global markets, helping them quickly conceptualize and address complex business data challenges.