Your company has Microsoft Sentinel deployed in multiple workspaces across several Azure regions. The security operations team wants to query data from all workspaces centrally using a single KQL query. What feature should you implement?
The workspace() expression in KQL—e.g., union workspace('Contoso-Sentinel').SecurityEvent, workspace('Fabrikam-Sentinel').SecurityEvent—lets you query tables across multiple Log Analytics workspaces in a single query. This is the Azure-native mechanism designed for Microsoft Sentinel multi-workspace scenarios and works without duplicating data or adding an extra ETL layer. It also supports resource-id based references, so you can query Sentinel workspaces that data is retained in while preserving the full context of the original table schema.
Why this answer
The workspace() expression in KQL allows you to include tables from multiple Log Analytics workspaces in a single query, enabling centralized querying across all Microsoft Sentinel workspaces without moving or aggregating data. This is the native and recommended approach for cross-workspace queries in Azure Sentinel.
Exam trap
The trap here is that candidates confuse the workspace() expression with the legacy linked workspace feature (Option A) or assume that a central aggregation mechanism (like a SIEM connector or data export) is required, when in fact KQL's native cross-workspace querying is the simplest and most cost-effective solution.
How to eliminate wrong answers
Option A is wrong because linked workspaces (via the legacy 'linked server' feature) are used for cross-region querying in Log Analytics but do not support the workspace() expression and are not designed for Microsoft Sentinel's multi-workspace architecture; they also require manual configuration and have limitations on the number of linked workspaces. Option C is wrong because exporting all data to Azure Data Explorer is unnecessary overhead—it introduces additional cost, latency, and complexity, and is not the standard method for querying Sentinel data across workspaces; the workspace() expression is simpler and more efficient. Option D is wrong because the Microsoft Sentinel SIEM connector is used to ingest data from external SIEMs (e.g., Splunk, ArcSight) into Sentinel, not to aggregate queries across multiple Sentinel workspaces.