Exam Bundle

An identified query and principal correlate with high resource consumption pertains to a different control (Query supported query-history or workload system data) and a different need (A team needs to attribute expensive SQL warehouse queries to their owners). It cannot establish that Aggregate recent system billing usage and apply a documented forecast assumption met the stated goal.

Aggregated usage by job or tag matches the reported consumption records would demonstrate the separate action Query the account's billing usage system tables and associated job metadata, which addresses “FinOps must allocate Databricks DBU consumption by job and cost center.” It does not verify the requirement “A team wants to forecast future monthly costs from observed daily spend” following Aggregate recent system billing usage and apply a documented forecast assumption.

Historical runs report success/failure and timing for the target job pertains to a different control (Use supported job run/system observability tables or Jobs API) and a different need (Operators need to track Lakeflow Jobs historical run states and durations). It cannot establish that Aggregate recent system billing usage and apply a documented forecast assumption met the stated goal.

Inspect filter pushdown and file/data skipping metrics in Query Profile belongs to the separate case “A scan reads far more rows than the WHERE clause should need,” where The scan indicates whether files or rows are being skipped as expected would be expected. That differs from the condition in this stem: “One join stage has an unexpectedly large shuffle and dominates the critical path.”

Use Spark UI or Query Profile partition/task timing to diagnose skew belongs to the separate case “A query is delayed by a small set of long-running skewed tasks,” where Outlier task durations and partition sizes identify the skewed key distribution would be expected. That differs from the condition in this stem: “One join stage has an unexpectedly large shuffle and dominates the critical path.”

Inspect query plan, filter selectivity and data layout metrics is appropriate when “A SQL query repeatedly scans a large table even for a narrow date predicate”; it would be verified by The scan plan shows poor pruning and the responsible predicate/layout. It does not directly satisfy “One join stage has an unexpectedly large shuffle and dominates the critical path,” whose relevant acceptance evidence is The join operator shows the largest shuffle and wall-clock contribution.

Outlier task durations and partition sizes identify the skewed key distribution pertains to a different control (Use Spark UI or Query Profile partition/task timing to diagnose skew) and a different need (A query is delayed by a small set of long-running skewed tasks). It cannot establish that Inspect Query Profile operators and shuffle bytes met the stated goal.

The plan reveals join strategy and whether broadcast thresholds were met would demonstrate the separate action Inspect the physical plan and build-side size estimates, which addresses “A join unexpectedly uses an expensive sort-merge stage instead of an eligible broadcast join.” It does not verify the requirement “One join stage has an unexpectedly large shuffle and dominates the critical path” following Inspect Query Profile operators and shuffle bytes.

The scan indicates whether files or rows are being skipped as expected would demonstrate the separate action Inspect filter pushdown and file/data skipping metrics in Query Profile, which addresses “A scan reads far more rows than the WHERE clause should need.” It does not verify the requirement “One join stage has an unexpectedly large shuffle and dominates the critical path” following Inspect Query Profile operators and shuffle bytes.

Use Spark UI or Query Profile partition/task timing to diagnose skew is appropriate when “A query is delayed by a small set of long-running skewed tasks”; it would be verified by Outlier task durations and partition sizes identify the skewed key distribution. It does not directly satisfy “A scan reads far more rows than the WHERE clause should need,” whose relevant acceptance evidence is The scan indicates whether files or rows are being skipped as expected.

The alternative Inspect the physical plan and build-side size estimates has a distinct purpose—A join unexpectedly uses an expensive sort-merge stage instead of an eligible broadcast join. It would not establish the required outcome (A scan reads far more rows than the WHERE clause should need); the necessary verification concerns The scan indicates whether files or rows are being skipped as expected.

The alternative Inspect Query Profile operators and shuffle bytes has a distinct purpose—One join stage has an unexpectedly large shuffle and dominates the critical path. It would not establish the required outcome (A scan reads far more rows than the WHERE clause should need); the necessary verification concerns The scan indicates whether files or rows are being skipped as expected.

The join operator shows the largest shuffle and wall-clock contribution would demonstrate the separate action Inspect Query Profile operators and shuffle bytes, which addresses “One join stage has an unexpectedly large shuffle and dominates the critical path.” It does not verify the requirement “A scan reads far more rows than the WHERE clause should need” following Inspect filter pushdown and file/data skipping metrics in Query Profile.

The scan plan shows poor pruning and the responsible predicate/layout pertains to a different control (Inspect query plan, filter selectivity and data layout metrics) and a different need (A SQL query repeatedly scans a large table even for a narrow date predicate). It cannot establish that Inspect filter pushdown and file/data skipping metrics in Query Profile met the stated goal.

This signal belongs to the adjacent case “A query is delayed by a small set of long-running skewed tasks,” whose remedy is Use Spark UI or Query Profile partition/task timing to diagnose skew. The observed result for the present requirement is The scan indicates whether files or rows are being skipped as expected.

Inspect query plan, filter selectivity and data layout metrics is appropriate when “A SQL query repeatedly scans a large table even for a narrow date predicate”; it would be verified by The scan plan shows poor pruning and the responsible predicate/layout. It does not directly satisfy “A query is delayed by a small set of long-running skewed tasks,” whose relevant acceptance evidence is Outlier task durations and partition sizes identify the skewed key distribution.

Inspect filter pushdown and file/data skipping metrics in Query Profile belongs to the separate case “A scan reads far more rows than the WHERE clause should need,” where The scan indicates whether files or rows are being skipped as expected would be expected. That differs from the condition in this stem: “A query is delayed by a small set of long-running skewed tasks.”

Inspect the physical plan and build-side size estimates is appropriate when “A join unexpectedly uses an expensive sort-merge stage instead of an eligible broadcast join”; it would be verified by The plan reveals join strategy and whether broadcast thresholds were met. It does not directly satisfy “A query is delayed by a small set of long-running skewed tasks,” whose relevant acceptance evidence is Outlier task durations and partition sizes identify the skewed key distribution.

This signal belongs to the adjacent case “A join unexpectedly uses an expensive sort-merge stage instead of an eligible broadcast join,” whose remedy is Inspect the physical plan and build-side size estimates. The observed result for the present requirement is Outlier task durations and partition sizes identify the skewed key distribution.

The scan plan shows poor pruning and the responsible predicate/layout pertains to a different control (Inspect query plan, filter selectivity and data layout metrics) and a different need (A SQL query repeatedly scans a large table even for a narrow date predicate). It cannot establish that Use Spark UI or Query Profile partition/task timing to diagnose skew met the stated goal.

That observation is a valid indicator for “A scan reads far more rows than the WHERE clause should need” when using Inspect filter pushdown and file/data skipping metrics in Query Profile. The scenario instead tests “A query is delayed by a small set of long-running skewed tasks”; the relevant result is Outlier task durations and partition sizes identify the skewed key distribution.

The alternative Use Spark UI or Query Profile partition/task timing to diagnose skew has a distinct purpose—A query is delayed by a small set of long-running skewed tasks. It would not establish the required outcome (A SQL query repeatedly scans a large table even for a narrow date predicate); the necessary verification concerns The scan plan shows poor pruning and the responsible predicate/layout.

This action solves a different issue: “A scan reads far more rows than the WHERE clause should need.” Its normal signal is The scan indicates whether files or rows are being skipped as expected. Here the requested behavior is “A SQL query repeatedly scans a large table even for a narrow date predicate,” so that signal is not sufficient.

This action solves a different issue: “A join unexpectedly uses an expensive sort-merge stage instead of an eligible broadcast join.” Its normal signal is The plan reveals join strategy and whether broadcast thresholds were met. Here the requested behavior is “A SQL query repeatedly scans a large table even for a narrow date predicate,” so that signal is not sufficient.

That observation is a valid indicator for “A scan reads far more rows than the WHERE clause should need” when using Inspect filter pushdown and file/data skipping metrics in Query Profile. The scenario instead tests “A SQL query repeatedly scans a large table even for a narrow date predicate”; the relevant result is The scan plan shows poor pruning and the responsible predicate/layout.

Outlier task durations and partition sizes identify the skewed key distribution pertains to a different control (Use Spark UI or Query Profile partition/task timing to diagnose skew) and a different need (A query is delayed by a small set of long-running skewed tasks). It cannot establish that Inspect query plan, filter selectivity and data layout metrics met the stated goal.

This signal belongs to the adjacent case “One join stage has an unexpectedly large shuffle and dominates the critical path,” whose remedy is Inspect Query Profile operators and shuffle bytes. The observed result for the present requirement is The scan plan shows poor pruning and the responsible predicate/layout.

This action solves a different issue: “One join stage has an unexpectedly large shuffle and dominates the critical path.” Its normal signal is The join operator shows the largest shuffle and wall-clock contribution. Here the requested behavior is “A join unexpectedly uses an expensive sort-merge stage instead of an eligible broadcast join,” so that signal is not sufficient.

The alternative Inspect query plan, filter selectivity and data layout metrics has a distinct purpose—A SQL query repeatedly scans a large table even for a narrow date predicate. It would not establish the required outcome (A join unexpectedly uses an expensive sort-merge stage instead of an eligible broadcast join); the necessary verification concerns The plan reveals join strategy and whether broadcast thresholds were met.

Use Spark UI or Query Profile partition/task timing to diagnose skew belongs to the separate case “A query is delayed by a small set of long-running skewed tasks,” where Outlier task durations and partition sizes identify the skewed key distribution would be expected. That differs from the condition in this stem: “A join unexpectedly uses an expensive sort-merge stage instead of an eligible broadcast join.”

This signal belongs to the adjacent case “A query is delayed by a small set of long-running skewed tasks,” whose remedy is Use Spark UI or Query Profile partition/task timing to diagnose skew. The observed result for the present requirement is The plan reveals join strategy and whether broadcast thresholds were met.

That observation is a valid indicator for “One join stage has an unexpectedly large shuffle and dominates the critical path” when using Inspect Query Profile operators and shuffle bytes. The scenario instead tests “A join unexpectedly uses an expensive sort-merge stage instead of an eligible broadcast join”; the relevant result is The plan reveals join strategy and whether broadcast thresholds were met.

This signal belongs to the adjacent case “A scan reads far more rows than the WHERE clause should need,” whose remedy is Inspect filter pushdown and file/data skipping metrics in Query Profile. The observed result for the present requirement is The plan reveals join strategy and whether broadcast thresholds were met.