A Brief History of Data Storage: Why OneLake?
Traditional Data Storage Methods
Organizations have stored data in many different ways throughout history. Early on, data lived on local servers or file systems. Then came the data warehouse era, which made it easier to analyze structured data through predefined schemas and optimized query engines. However, data warehouses were not built for unstructured data such as images, videos, or raw text files.
In the early 2010s, the concept of the data lake emerged. A data lake could store any type of data — structured, semi-structured, or unstructured — in its raw form. Yet data lakes brought their own set of problems: data quality was difficult to enforce, governance was nearly impossible to maintain, and many organizations ended up with what practitioners called a "data swamp."
OneLake combines the flexibility of a data lake with the discipline of a data warehouse, representing a new generation of data storage. Just as OneDrive does for personal files, OneLake does for enterprise data: it stores everything in a centralized, organized repository and makes it instantly accessible whenever needed.
Definition and Core Concept
OneLake is the central data storage layer of Microsoft Fabric. All organizational data is stored in OneLake, and every Fabric workload — Data Engineering, Data Science, Real-Time Analytics, and Power BI — reads from and writes to this single store.
OneLake's core philosophy is simple: Data is stored once and used many times. Once a data engineer loads data into OneLake, analysts can query it, data scientists can train machine learning models on it, and business users can build reports from it. Everyone works from the same data, which guarantees consistency across the entire organization.
OneLake is built on top of Microsoft Azure's cloud infrastructure. It stores data on Azure Data Lake Storage (ADLS) and uses the Delta Lake open format. Delta Lake provides critical advantages in terms of data reliability, ACID transactions, and time travel capabilities.
OneLake's architecture consists of three main layers:
Data Layer: The physical layer where raw data is stored. This is where files and tables actually reside.
Metadata Layer: The layer that stores information about the data — data types, sizes, creation dates, schema definitions, and more. This layer enables fast discovery and organization of data assets.
Access Layer: The layer through which users and applications interact with data. Security controls, permissions, and authentication are enforced here.
Centralized Data Repository
OneLake serves as the single centralized store for all organizational data. Data from Marketing, Finance, Operations, and other departments is organized within OneLake, eliminating data silos that slow down decision-making.
OneLake scales seamlessly to petabyte-level data volumes. As the organization grows and data volumes increase, OneLake scales automatically. There are no storage limitations to worry about.
Data Security and Governance
OneLake follows enterprise-grade security standards. Access to data is managed through Role-Based Access Controls (RBAC) and sensitivity labels. Additionally, integration with Microsoft Purview ensures comprehensive data governance and regulatory compliance.
OneLake's Direct Lake mode allows Power BI reports to access data in OneLake directly, without importing or caching. This dramatically improves report loading speeds and enables true real-time analysis.
The Evolution of the Data Lake Concept
From Data Warehouse to Data Lake
A data warehouse is ideal for structured data with predefined schemas. However, modern organizations work with every type of data: social media feeds, IoT sensor readings, video files, audio recordings, and more. A traditional data warehouse is simply not designed to handle this variety.
The data lake was created to solve this problem. A data lake can store any type of data without requiring a schema upfront. However, as mentioned, data lakes often degraded into data swamps where data quality and discoverability were poor.
Lakehouse: The Best of Both Worlds
The Lakehouse architecture — built on top of OneLake — combines the flexibility of a data lake with the structure and reliability of a data warehouse. A Lakehouse can store any type of data (like a data lake) while also enforcing data quality, schema management, and ACID transactions (like a data warehouse). OneLake is the foundation upon which this Lakehouse architecture is built.
Data Organization in OneLake
In OneLake, data is organized within workspaces. Each workspace can represent a separate department, project, or business unit. For example, a "Marketing Analytics" workspace can contain all data assets belonging to the marketing department.
To create a workspace, navigate to the Fabric portal and select "Workspaces" from the left-hand menu. Click "+ New Workspace," enter a name for the workspace, and configure the required permissions.
Within a workspace, data can be further divided into folders and domains. For example, within the "Marketing Analytics" workspace, you might create separate folders for "Customer Data," "Campaign Data," and "Sales Data."
To create a folder, click "+ New Folder" within the workspace, enter a name, and add a description. Folders make it straightforward to organize data assets in a logical hierarchy.
Momentum Datalabs advocates for a data mesh approach to data organization. Data mesh distributes data ownership to individual departments, making each team responsible for the quality and governance of their own data. OneLake is the ideal platform for implementing this approach.
Under a data mesh model, each department:
•Organizes its own data within OneLake
•Takes ownership of data quality
•Shares data with other departments through governed interfaces
•Adheres to enterprise-wide data governance principles
Data Virtualization in OneLake: Shortcuts
In OneLake, you can access data in other locations without physically copying it. The mechanism that makes this possible is called shortcuts.
Shortcuts work like symbolic links. For example, the Marketing department's customer data lives in its own workspace. When the Finance department needs access to that data, instead of creating a physical copy, they can create a shortcut. Finance can then access the customer data directly, while the data itself remains in a single location.
Eliminates Data Duplication: The same data is never stored in multiple places. This reduces storage costs and guarantees data consistency across the organization.
Clarifies Data Ownership: Data is managed by the team that owns it. Other teams access it through shortcuts, preserving clear ownership boundaries.
Provides Flexibility: When the data structure changes, there is no need to update multiple copies. The shortcut automatically reflects the latest version of the data.
Step-by-Step Guide to Creating Shortcuts
Follow the steps below to create a shortcut in OneLake:
Step 1: Open Your Lakehouse
Navigate to your workspace in the Fabric portal and open the lakehouse in which you want to create the shortcut.
Step 2: Right-Click to Open the Context Menu
In the Explorer pane, right-click on the folder where you want to create the shortcut — for example, under the "Files" folder or the "Tables" folder.
Step 3: Select "New Shortcut"
From the context menu, select New shortcut, New table shortcut, or New schema shortcut, depending on your lakehouse configuration.
Step 4: Choose a Data Source
In the dialog that opens, select Microsoft OneLake under "Internal sources." Then select the data source you want to connect to (a lakehouse, warehouse, etc.) and click Next.
Expand the "Files" or "Tables" folders within the data source. Folders containing Delta or Iceberg format tables are indicated with a table icon; other folders are shown with a standard folder icon.
Select one or more folders to connect to (up to 50 folders at a time) and click Next.
Step 6: Create the Shortcut
Review your selected shortcut locations. Use the edit action to change the default shortcut name if needed, and the delete action to remove any unwanted selections. When ready, click Create.
Once the shortcut is created, it will appear under the selected directory in the Explorer pane and can be used just like a regular folder.
To edit or delete an existing shortcut:
1.Right-click on the shortcut
3.Edit the shortcut name, target connection, or target location as needed
To delete a shortcut, right-click and select Delete. Deleting a shortcut does not affect the original data in the target location.
Data Security and Access Controls in OneLake
Role-Based Access Controls (RBAC)
In OneLake, access to data is controlled based on the user's assigned role. For example, a data engineer may have write permissions, while an analyst has read-only access.
The primary roles in OneLake are as follows:
Admin: Can manage the workspace and all its data assets. Can configure user permissions.
Member: Can create, edit, and delete data within the workspace.
Contributor: Can create and edit data within the workspace, but cannot delete items.
Viewer: Can only view data within the workspace; no write access.
To assign roles, navigate to the workspace settings and click "Access." Add users and assign the appropriate role.
In OneLake, you can apply sensitivity labels to data assets. For example, customer personal data can be labeled "Highly Confidential." This label governs how the data is handled and who is permitted to access it.
To apply a sensitivity label:
1.Right-click on the data asset
3.Choose the appropriate label (Public, Internal, Confidential, or Highly Confidential)
Data Governance in OneLake
Microsoft Purview Integration
Microsoft Purview is the enterprise data governance and compliance platform that integrates natively with OneLake. Through Purview, organizations can catalog all data assets, track data lineage, enforce data quality rules, and generate compliance reports.
With the Purview hub, you can:
•Build a comprehensive catalog of all data assets across OneLake
•Track where data originates and how it has been transformed
•Define and enforce data quality rules
•Generate regulatory compliance reports
Data Cataloging and Labeling
In OneLake, data assets are automatically cataloged and tagged. For example, customer data assets receive a "Customer Data" tag. This makes it easy to discover and organize data across the entire organization.
To access the data catalog, click "Data Hub" in the Fabric portal. Here you will find a complete list of all data assets stored in OneLake.
Momentums' OneLake Strategy
Designing Scalable Data Platforms
Momentum Datalabs designs scalable, OneLake-based data platforms tailored to each organization's specific requirements. The data architecture is planned to grow alongside the business, ensuring that the platform remains performant and cost-efficient as data volumes increase.
Momentum Datalabs applies the data mesh approach to distribute data ownership across departments. Each team takes responsibility for its own data, while governed sharing mechanisms ensure that cross-departmental collaboration remains seamless.
Data Governance and Security
Momentum Datalabs applies industry best practices for data security and governance within OneLake. Through sensitivity labels, access controls, and Microsoft Purview integration, enterprise data is protected and compliance requirements are consistently met.
A Real-World Example: OneLake in Action
Scenario: A Multi-Departmental Organization
Consider an e-commerce company with Marketing, Sales, Operations, and Finance departments. In a traditional setup, each department operates its own data system:
•Marketing: Stores customer segmentation and campaign data in its own system
•Sales: Stores transaction and customer interaction data in a separate system
•Operations: Stores inventory and logistics data in yet another system
•Finance: Stores accounting and budget data in a separate system
This fragmentation creates data silos and slows down decision-making across the organization.
With OneLake, all of this data is organized within a single centralized repository:
1.Marketing Workspace: Customer segmentation and campaign data
2.Sales Workspace: Sales transactions and customer interactions
3.Operations Workspace: Inventory and logistics data
4.Finance Workspace: Accounting and budget data
Each department manages its own data. Other departments access it through shortcuts, without creating redundant copies. For example, the Finance team can access Sales data via a shortcut and build financial reports that incorporate real-time sales figures.
Because all data lives in OneLake, Power BI reports can combine data from all departments to deliver a unified, organization-wide view.
Conclusion: Achieving a Single Source of Truth
OneLake represents a fundamental shift in how organizations manage their data. A centralized data repository eliminates data silos and delivers a true Single Source of Truth. Momentum leverages the power of OneLake to guide organizations through the transition to scalable, secure, and well-governed data platforms.
Unlocking the full potential of data starts with organizing it and consolidating it in a single place. OneLake is the first step on that journey.
Partner with Momentum to organize all your enterprise data in OneLake and achieve a Single Source of Truth.
Microsoft Fabric
18.04.2026
