DP-600 - Prepare Data (46% of the exam) - Section 2.8

Select, filter, and aggregate data using KQL against an Eventhouse or KQL database.

Write Kusto Query Language (KQL) queries using the summarize operator, time-series functions, and filters to analyse data stored in an Eventhouse or KQL database. Recognise when KQL is the appropriate query tool compared to T-SQL, particularly for high-frequency telemetry and time-series workloads.

Kusto Query LanguageEventhouseKQL querysetsummarize operatortime-series queries

Practice question for this objective

Free samplePrepare Datamedium

An analytics engineer in "Microsoft Fabric" wants to author, save, and share a set of exploratory KQL queries that run against tables in an "Eventhouse", keeping the queries grouped as reusable tabs alongside the data. Which Fabric item is designed to hold and run these saved KQL queries?

AA KQL queryset, which stores multiple named KQL queries as tabs that run against Eventhouse and KQL database tables and can be saved and shared as a Fabric item. Correct
BA "Dataflows Gen2" item, which authors the reusable transformations and saves them as steps that run against the Eventhouse tables for the exploratory analysis required.
CA "SQL analytics endpoint", which stores the reusable query definitions as views that run against the Eventhouse tables and can be saved and shared as a Fabric item.
DA "semantic model", which stores the reusable query definitions as measures that run against the Eventhouse tables and can be saved and shared as a Fabric item.

Identify the KQL queryset as the Fabric item for authoring, organising, saving, and sharing KQL queries against Eventhouse and KQL database data. A KQL queryset is the dedicated Fabric authoring tool for Kusto Query Language: it groups multiple queries into named tabs and runs them against Eventhouse and KQL database tables, which neither dataflows, semantic models, nor SQL endpoints provide.

Why A is correct: A KQL queryset is the Fabric item purpose-built to author, organise into tabs, save, and share KQL queries that run directly against Eventhouse and KQL database data.

Why B is wrong: Dataflows Gen2 uses Power Query for ingestion and transformation, not KQL authoring, so it is the wrong surface for writing and saving Kusto queries against an Eventhouse.

Why C is wrong: A SQL analytics endpoint exposes a read-only T-SQL surface over Lakehouse tables, not a KQL authoring workspace, so it cannot hold tabbed Kusto querysets.

Why D is wrong: A semantic model defines DAX measures and relationships for reporting, not saved KQL queries, so it does not serve as the authoring surface for Kusto exploration.

See more DP-600 practice questions, answers explained.

Exam traps in Prepare Data

Answers that look right on this material and are not. Each one is a distractor from a different question in the DP-600 bank for this domain.

A "Warehouse", which stores the streaming telemetry in tables exposed for read-write T-SQL so the engineer can query it with Kusto Query Language and time-series operators.
Why it is wrong: A Warehouse is a T-SQL read-write store and does not run Kusto Query Language, so it cannot serve the KQL and time-series querying the requirement specifies.
A "Warehouse", because its T-SQL engine is purpose-built for high-cardinality time-series metrics and interactive time-window aggregation over months of readings.
Why it is wrong: A "Warehouse" handles structured relational data but is not the engine optimised for high-cardinality time-series and log exploration, so it is not the purpose-built fit described.
A "Lakehouse", because its Delta tables ingest streaming telemetry continuously and its query surface is tuned for sub-second time-series aggregation over billions of rows.
Why it is wrong: A "Lakehouse" suits batch and Spark engineering over Delta tables, and it is not optimised for continuous low-latency event ingestion or the time-series query language the workload requires.

Examworthy is not affiliated with or endorsed by Microsoft. Original, blueprint-aligned practice material only.