Microsoft Sentinel is leveling up! Already a trusted cloud-native Security Information and Event Management (SIEM) and Security Orchestration, Automation and Response (SOAR) solution, it empowers security teams to detect, investigate, and respond to threats with speed and precision. Now, with the introduction of its new Data Lake architecture, Sentinel is transforming how security data is stored, accessed, and analyzed, bringing unmatched flexibility and scale to threat investigation.

Unlike Microsoft Fabric OneLake, which supports analytics across the organization, Sentinel’s Data Lake is purpose-built for security. It centralizes raw structured, semi-structured, and unstructured data in its original format, enabling advanced analytics without rigid schemas.

This article is written by someone who’s spent years helping security teams navigate Microsoft’s evolving ecosystem, translating complex capabilities into practical strategies. What follows is a hands-on look at the key features, benefits, and challenges of Sentinel’s Data Lake, designed to help you make the most of this powerful new architecture.

Current Sentinel Features

To tackle the challenges security teams, face today—like explosive data growth, integration of varied sources, and tight compliance requirements—organizations need scalable, efficient architectures. Legacy SIEMs often become costly and slow when analyzing multi-year data or correlating diverse events. Security data lakes address these issues by enabling seamless ingestion of logs from any source, schema-on-read flexibility, and parallelized queries over massive datasets. This schema–on-read allows SOC analysts to define how data is interpreted at the time of analysis, rather than when it is stored. This means analysts can flexibly adapt queries and threat detection logic to evolving threats, without reformatting historical data making investigations more agile and responsive to change.

This empowers security operations to conduct deep historical analysis, automate enrichment, and apply advanced analytics, such as machine learning, while retaining strict control over data access and residency. Ultimately, decoupling storage and compute allows teams to boost detection and response speed, maintain compliance, and adapt their Security Operation Center (SOC) to future security demands.

As organizations manage increasing data and limited budgets, many are moving from legacy SIEMs to advanced cloud-native options. Microsoft Sentinel’s Data Lake separates storage from computing, offering scalable and cost-effective analytics and compliance. For instance, storing 500 TB of logs in Sentinel Data Lake can cut costs by 60–80% compared to Log Analytics, due to lower storage costs and flexible retention. Integration with modern tools and open formats enables efficient threat response and regulatory compliance. Microsoft Sentinel data lake pricing (preview)

Sentinel Data Lake Use Cases

• Log Retention: Long-term retention of security logs for compliance and forensic investigations


• Hunting: Advanced threat hunting using historical data


• Interoperability: Integration with Microsoft Fabric and other analytics platforms


• Cost: Efficient storage prices for high-volume data sources

How Microsoft Sentinel Data Lake Helps

Microsoft Sentinel’s Data Lake introduces a powerful paradigm shift for security operations by architecting the separation of storage and compute, enabling organizations to achieve petabyte-scale data retention without the traditional overhead and cost penalties of legacy SIEM solutions. Built atop highly scalable, cloud-native infrastructure, Sentinel Data Lake empowers SOCs to ingest telemetry from virtually unlimited sources ranging from on-premises firewalls, proxies, and endpoint logs to SaaS, IaaS, and PaaS environments—while leveraging schema-on-read, a method that allows analysts to define how data is interpreted at query time rather than when it is stored, offering greater flexibility in analytics. For example, a security analyst can adapt to the way historical data is examined as new threats emerge, without needing to reformat or restructure the data stored in the Data Lake.

From Microsoft Learn – Retention and data tiering

Storing raw security logs in open formats like Parquet (this is a columnar storage file format optimized for efficient data compression and retrieval, commonly used in big data processing frameworks like Apache Spark and Hadoop) enables easy integration with analytics tools and Microsoft Fabric, letting analysts efficiently query historical data using KQL, SQL, or Spark. This approach eliminates the need for complex ETL and archived data rehydration, making incident response faster; for instance, a SOC analyst can quickly search for years of firewall logs for threat detection.

From Microsoft Learn – Flexible querying with Kusto Query Language

Granular data governance and access controls allow organizations to manage sensitive information and meet legal requirements. Storing raw security logs in open formats enables fast investigations of long-term data incidents, while automated lifecycle management reduces costs and ensures compliance. Data Lakes integrate with Microsoft platforms and other tools for unified analytics and security. Machine learning helps detect unusual login activity across years, overcoming previous storage issues.

From Microsoft Learn – Powerful analytics using Jupyter notebooks

Pros and Cons

The following table highlights the advantages and potential opportunities that Microsoft Sentinel Data Lake offers. This follows the same Pay-As-You-Go pricing model as currently available with Sentinel.