How BlueCloud helped a global reseller transform complex Microsoft Fabric data pipelines into a scalable Snowflake-driven solution
The journey from scattered enterprise data to actionable intelligence can be filled with hidden obstacles—especially when systems like Microsoft Fabric and Dynamics 365 (D3 65) are involved.
For a global reseller seeking to unify and modernize their data pipeline, this challenge was all too familiar. Their existing setup based on Microsoft’s ecosystem was cumbersome, lacked flexibility, and failed to deliver the speed and reliability the business needed. They turned to BlueCloud for a solution that would allow them to fully harness the power of the Snowflake Data Cloud while maintaining seamless integration with Microsoft Fabric.
What followed was a story of engineering precision, creative problem-solving, and the strategic use of Snowflake’s architectural strengths to overcome platform limitations and unlock new business potential.
The Challenge: Bridging Microsoft Fabric and Snowflake
When BlueCloud’s Principal Data Engineer, Konstantin Morozov, first joined the project, he met with a seemingly simple but technically demanding request: migrated the pipeline from SQL Server to Snowflake, ensuring minimal data movement and maintaining synchronization between systems.
However, as Morozov quickly discovered, no native connector existed between Dynamics 365 (Data 65) and Snowflake.
“The challenge was to bring the data out of D365 and put it into Snowflake with minimal data movement and minimal friction between the two systems,” explains Konstantin Morozov. “There was no native connector on either side that could be utilized for this.”
Initially, Microsoft representatives recommended the use of Microsoft Fabric, as it provided a native integration path from Dynamics 365 through Azure Synapse Link for Dataverse service or simply Dataverse Link. Fabric, they suggested, could act as a bridge, though it wasn’t designed to connect directly with Snowflake.
While Fabric offered an easier way to pull data from Dynamics 365, it introduced a new set of limitations: restricted configuration options, a lack of granular control over synchronization, and an opaque data management model.
“Dataverse Link is a Microsoft-managed service,” Morozov explains. “It’s a one-button setup. You flip the switch, and the tables are synchronized between Data 65 and Fabric. But there’s not much control over the process, and that creates challenges when you want to build a robust, scalable data architecture.”
Finding the Breakthrough: Harnessing Snowflake’s Iceberg Capability
The first step was to get data from Microsoft Fabric’s Delta Lake into Snowflake without creating excessive overhead or duplication. The goal: minimize data movement and storage costs while enabling real-time or near-real-time visibility for business intelligence and analytics.
BlueCloud’s engineering team explored multiple options to connect the two platforms. One approach was to use Azure Data Factory (ADF) pipelines to extract and load data from Fabric into Snowflake. While technically feasible, this method proved costly and inefficient, particularly given the large number of tables (hundreds of them, with some containing 100-200GB of data).
Running these pipelines multiple times per day would have led to unsustainable compute and storage costs. That’s when the team looked deeper into what Snowflake could offer beyond traditional ETL or ELT workflows.
“What we realized was that Snowflake’s support for Iceberg tables opened a better way to connect,” says Morozov. “Instead of physically moving data from Fabric to Snowflake, we could create an abstraction layer using Iceberg tables that allowed Snowflake to read Delta tables directly.”
This approach became the foundation of the solution. By using Snowflake Iceberg tables on top of Microsoft Fabric’s Delta tables, BlueCloud effectively eliminated the need to transfer data between systems. Instead, Snowflake acted as a query layer, directly accessing Fabric data through synchronized metadata.
The benefits:
- Zero duplication: Data remained in Fabric, reducing storage requirements.
- Faster access: Snowflake could query live data without delay.
- Lower maintenance: No need for continuous pipeline updates or costly data transfers.
- Improved scalability: The setup could easily handle hundreds of tables with minimal operational overhead.
Building a Connected, High-Performance Data Ecosystem
Once the connection was established, BlueCloud fine-tuned the architecture to ensure reliability, scalability, and data freshness. The solution following technologies:
- Snowflake for scalable data access and analytics
- Microsoft Fabric for native integration with Dynamics 365
- Sigma for advanced reporting and dashboard creation
- Python for lightweight custom scripting and automation
- Coalesce for building the transformation pipeline
“We wanted to keep the architecture as clean as possible,” Morozov notes. “The client was using multiple tools, and the goal was to move as much as possible into Snowflake. Snowflake’s unified approach to compute and storage simplifies everything, data modeling, analytics, and even performance optimization.”
This decision reflected a growing trend among BlueCloud’s clients: centralizing their data operations on Snowflake to take advantage of its flexibility, security, and ecosystem integration.
Snowflake’s separation of compute and storage, support for open formats like Iceberg, and ability to integrate seamlessly with analytics tools like Sigma make it an ideal foundation for modern data strategies.
The Business Impact: Real-Time Visibility and Better Decisions
The result was not just a technical success—it had a direct impact on the client’s operations. By establishing a near-real-time data flow from Dynamics 365 to Snowflake, BlueCloud helped the client generate live dashboards that visualize key business metrics such as orders, inventory levels, and product performance.
These dashboards, built using Sigma and powered by Snowflake’s query engine, gave both the client and their partners instant visibility into their most critical data.
“The client can now see up-to-date metrics across orders, inventory, and products. That level of visibility just wasn’t possible with their old setup,” says Morozov.
This shift allowed business teams to make data-backed decisions faster, adjust stock, optimize logistics, and respond to customer trends with greater agility. For the client’s customers, the dashboards enabled more accurate forecasting and smoother supply chain collaboration.
From a business perspective, the integration improved efficiency and reduced costs by minimizing data movement and duplication.
“We achieved a balance between performance, cost, and flexibility,” explains Morozov. “By using Snowflake as the analytical layer and Fabric as the data source, we delivered a solution that’s both efficient and easy to maintain.”
Ongoing Optimization: Continuous Improvement and Future Readiness
The project remains active as BlueCloud continues to refine and optimize the pipeline. With the foundational connection between Fabric and Snowflake stabilized, the team’s focus has shifted toward enhancing business logic and performance tuning within the Snowflake environment.
“Now that the connection is stable and the pipeline runs smoothly, we are focusing more on improving the business logic,” says Morozov. “It’s a giant system, and we’re constantly refining it to make it faster and more efficient.”
This continuous improvement mindset reflects BlueCloud’s broader philosophy: building adaptive data ecosystems that evolve alongside client needs.
As more organizations make Snowflake their central data hub, solutions like this, which connect modern platforms like Microsoft Fabric, show how to leverage hybrid cloud environments while keeping control and scalability intact.
Why Snowflake Matters
At the heart of this transformation is Snowflake’s architecture, which allows organizations to break down silos and operate with agility across diverse data sources.
Here’s why Snowflake plays such a crucial role in overcoming challenges like these:
- Unified Data Platform:
Snowflake allows seamless integration of structured, semi-structured, and unstructured data in a single environment, eliminating the fragmentation that often plagues enterprise systems.
- Separation of Compute and Storage:
This enables cost efficiency and scalability, as compute power can scale independently based on workloads—ideal for fluctuating data ingestion and analytics demands.
- Native Support for Open Table Formats (Iceberg, Delta):
Snowflake’s support for open formats enables direct queries on external data sources—exactly the breakthrough that made the Fabric integration possible in this project.
- Secure Data Sharing:
Snowflake simplifies controlled data sharing, allowing organizations to collaborate across regions and partners without duplicating datasets.
- Performance and Reliability:
With near-infinite concurrency and auto-scaling capabilities, Snowflake provides consistent performance even with massive datasets—crucial for real-time dashboards and business intelligence.
Turning Complexity into Clarity with BlueCloud
For BlueCloud, projects like this exemplify their mission: to help clients simplify complexity and turn fragmented data into meaningful insights. By combining engineering expertise with business-focused thinking, the team ensures that every solution delivers measurable outcomes.
“It’s not just about moving data. It’s about enabling better business outcomes,” Morozov emphasizes. “When clients can see their data clearly, they can act faster, plan smarter, and grow more confidently.”
BlueCloud’s experience working across cloud ecosystems, Snowflake, Azure, AWS, and beyond, enables them to design architectures that aren’t just technically sound but also strategically aligned with client goals.
“The beauty of Snowflake,” concludes Morozov, “is that it gives you the flexibility to work with any source, any format, and any scale. Once you understand how to unlock that power, you can turn even the most complex data environments into something elegant and efficient.”
Explore our success stories to learn more about how we help organizations deliver Enterprise AI, through high-performing and scalable global teams.
