Repair a report plan without changing its query results
A hands-on PL-300 lab. You produce the real artefact and 9 automated checks verify it behaves the way the exam expects.
Try this labAll PL-300 practice
- Certification
- PL-300
- Format
- Structured configuration
- Difficulty
- hard
- Estimated time
- 35 min
- Automated checks
- 9
The brief
Edit performance.plan.json. projection lists retained source fields; indexes lists equality-key field lists; aggregate is null or one cache with groupBy and reducers keyed by Revenue/ObservationCount. Reducers use operation count-rows, or operation sum/average/max/count-nonblank plus a projected column. Revenue's required result is the sum of Amount; ObservationCount counts all rows. A cache must cover every query grouping/filter field and rendered metric to be used, otherwise a detail index/full scan supplies the query. Indexes require all key filters and no more than 128 lookup combinations. The runtime chooses the smallest eligible candidate source and sums cached metrics when regrouping. visualMeasures maps every query ID to its rendered metric list; preserve requested measures and remove unnecessary ones. rowLimit is null or an integer 0..128; result truncation is tested against every complete result. Stored units count projected values, cache keys/metric values and index keys/references; diagnostic source IDs are excluded instrumentation. Row visits count candidate and retained records. Rendered cells count output groups times grouping-field count plus rendered-measure count. Keep region/year/category context, exact detail identities and zero/null/empty semantics. Inspect independent result and cost receipts, then repair without Reset.
What the checks verify
Your work is graded on 9 independent properties, not on matching one reference answer.
- Compile the declared projection, index keys, reducers and visual measure bindings.
- Monthly revenue preserves every contributing source identity and the correct grouped sums.
- Category counts include every original observation, including null-amount rows.
- Other-region and prior-year queries preserve their own grouping and filtered populations.
- Two individual observation queries retain exact source detail and values rather than summary rows.
- Zero, null amount, absent member and empty context retain their distinct observations and aggregates.
- Every query visits at most seventy candidate-plus-retained rows under the explicit local work model.
- Projected values, cached values and index keys/references fit nine hundred local memory units.
- Visible grouping cells plus rendered measures fit each query's complete-result cell budget.
Where this sits in the PL-300 blueprint
- Domain
- Model the Data
- Objective
- Optimize Model Performance
- Skill
- Analyzer and Query Evidence
Part of PL-300 preparation
Labs are written by ExamNova to teach the decisions the exam tests. They are not reproductions of vendor lab content.