Exam Bundle

This signal belongs to the adjacent case “Queries spend excessive time opening small files after frequent appends,” whose remedy is Run `OPTIMIZE` or enable supported automatic compaction. The observed result for the present requirement is Queries use enough scan tasks without excessive file-open overhead.

This signal belongs to the adjacent case “A daily writer creates millions of tiny partitions from a high-cardinality identifier,” whose remedy is Remove pathological partitioning and redesign data layout. The observed result for the present requirement is Queries use enough scan tasks without excessive file-open overhead.

Reads see consistent table versions before and after compaction would demonstrate the separate action Use supported Delta OPTIMIZE rather than directly rewriting the data directory, which addresses “A table needs to preserve transaction correctness during physical file compaction.” It does not verify the requirement “Very large files cause poor parallelism for typical query scans” following Choose a balanced target file size and tune supported write optimization.

This action solves a different issue: “A query wraps a partition/filter column in a nonpushdown-compatible function.” Its normal signal is Physical plan pushes the predicate and scans fewer files. Here the requested behavior is “A query filters by an indexed or statistically tracked customer/date field,” so that signal is not sufficient.

Add a semantically correct selective WHERE condition to the query is appropriate when “The table's data is well laid out but a query still performs full scans because no filter was supplied”; it would be verified by The plan shows pruning and returned records satisfy the intended subset. It does not directly satisfy “A query filters by an indexed or statistically tracked customer/date field,” whose relevant acceptance evidence is Scan metrics show a lower percentage of files read.

Use liquid clustering or suitable layout instead of one partition per key belongs to the separate case “A high-cardinality filter benefits from clustering rather than massive Hive partitions,” where File count and selective scan cost decrease without partition explosion would be expected. That differs from the condition in this stem: “A query filters by an indexed or statistically tracked customer/date field.”

File count and selective scan cost decrease without partition explosion would demonstrate the separate action Use liquid clustering or suitable layout instead of one partition per key, which addresses “A high-cardinality filter benefits from clustering rather than massive Hive partitions.” It does not verify the requirement “A query filters by an indexed or statistically tracked customer/date field” following Ensure relevant column statistics and appropriate data layout for pruning.

This signal belongs to the adjacent case “The table's data is well laid out but a query still performs full scans because no filter was supplied,” whose remedy is Add a semantically correct selective WHERE condition to the query. The observed result for the present requirement is Scan metrics show a lower percentage of files read.

This signal belongs to the adjacent case “A query wraps a partition/filter column in a nonpushdown-compatible function,” whose remedy is Rewrite the predicate to a sargable equivalent while preserving semantics. The observed result for the present requirement is Scan metrics show a lower percentage of files read.

This action solves a different issue: “A query wraps a partition/filter column in a nonpushdown-compatible function.” Its normal signal is Physical plan pushes the predicate and scans fewer files. Here the requested behavior is “A Delta table lacks statistics for a frequently filtered column,” so that signal is not sufficient.

The alternative Add a semantically correct selective WHERE condition to the query has a distinct purpose—The table's data is well laid out but a query still performs full scans because no filter was supplied. It would not establish the required outcome (A Delta table lacks statistics for a frequently filtered column); the necessary verification concerns The optimizer reports effective pruning on that predicate.

The alternative Use liquid clustering or suitable layout instead of one partition per key has a distinct purpose—A high-cardinality filter benefits from clustering rather than massive Hive partitions. It would not establish the required outcome (A Delta table lacks statistics for a frequently filtered column); the necessary verification concerns The optimizer reports effective pruning on that predicate.

This signal belongs to the adjacent case “A query wraps a partition/filter column in a nonpushdown-compatible function,” whose remedy is Rewrite the predicate to a sargable equivalent while preserving semantics. The observed result for the present requirement is The optimizer reports effective pruning on that predicate.

That observation is a valid indicator for “A high-cardinality filter benefits from clustering rather than massive Hive partitions” when using Use liquid clustering or suitable layout instead of one partition per key. The scenario instead tests “A Delta table lacks statistics for a frequently filtered column”; the relevant result is The optimizer reports effective pruning on that predicate.

Scan metrics show a lower percentage of files read pertains to a different control (Ensure relevant column statistics and appropriate data layout for pruning) and a different need (A query filters by an indexed or statistically tracked customer/date field). It cannot establish that Collect or configure supported data skipping statistics for that column met the stated goal.

Use liquid clustering or suitable layout instead of one partition per key belongs to the separate case “A high-cardinality filter benefits from clustering rather than massive Hive partitions,” where File count and selective scan cost decrease without partition explosion would be expected. That differs from the condition in this stem: “A query wraps a partition/filter column in a nonpushdown-compatible function.”

Ensure relevant column statistics and appropriate data layout for pruning belongs to the separate case “A query filters by an indexed or statistically tracked customer/date field,” where Scan metrics show a lower percentage of files read would be expected. That differs from the condition in this stem: “A query wraps a partition/filter column in a nonpushdown-compatible function.”

This action solves a different issue: “The table's data is well laid out but a query still performs full scans because no filter was supplied.” Its normal signal is The plan shows pruning and returned records satisfy the intended subset. Here the requested behavior is “A query wraps a partition/filter column in a nonpushdown-compatible function,” so that signal is not sufficient.

File count and selective scan cost decrease without partition explosion pertains to a different control (Use liquid clustering or suitable layout instead of one partition per key) and a different need (A high-cardinality filter benefits from clustering rather than massive Hive partitions). It cannot establish that Rewrite the predicate to a sargable equivalent while preserving semantics met the stated goal.

Scan metrics show a lower percentage of files read would demonstrate the separate action Ensure relevant column statistics and appropriate data layout for pruning, which addresses “A query filters by an indexed or statistically tracked customer/date field.” It does not verify the requirement “A query wraps a partition/filter column in a nonpushdown-compatible function” following Rewrite the predicate to a sargable equivalent while preserving semantics.

That observation is a valid indicator for “A Delta table lacks statistics for a frequently filtered column” when using Collect or configure supported data skipping statistics for that column. The scenario instead tests “A query wraps a partition/filter column in a nonpushdown-compatible function”; the relevant result is Physical plan pushes the predicate and scans fewer files.

Add a semantically correct selective WHERE condition to the query is appropriate when “The table's data is well laid out but a query still performs full scans because no filter was supplied”; it would be verified by The plan shows pruning and returned records satisfy the intended subset. It does not directly satisfy “A high-cardinality filter benefits from clustering rather than massive Hive partitions,” whose relevant acceptance evidence is File count and selective scan cost decrease without partition explosion.

This action solves a different issue: “A query filters by an indexed or statistically tracked customer/date field.” Its normal signal is Scan metrics show a lower percentage of files read. Here the requested behavior is “A high-cardinality filter benefits from clustering rather than massive Hive partitions,” so that signal is not sufficient.

Collect or configure supported data skipping statistics for that column belongs to the separate case “A Delta table lacks statistics for a frequently filtered column,” where The optimizer reports effective pruning on that predicate would be expected. That differs from the condition in this stem: “A high-cardinality filter benefits from clustering rather than massive Hive partitions.”

That observation is a valid indicator for “A query wraps a partition/filter column in a nonpushdown-compatible function” when using Rewrite the predicate to a sargable equivalent while preserving semantics. The scenario instead tests “A high-cardinality filter benefits from clustering rather than massive Hive partitions”; the relevant result is File count and selective scan cost decrease without partition explosion.

That observation is a valid indicator for “The table's data is well laid out but a query still performs full scans because no filter was supplied” when using Add a semantically correct selective WHERE condition to the query. The scenario instead tests “A high-cardinality filter benefits from clustering rather than massive Hive partitions”; the relevant result is File count and selective scan cost decrease without partition explosion.

Scan metrics show a lower percentage of files read would demonstrate the separate action Ensure relevant column statistics and appropriate data layout for pruning, which addresses “A query filters by an indexed or statistically tracked customer/date field.” It does not verify the requirement “A high-cardinality filter benefits from clustering rather than massive Hive partitions” following Use liquid clustering or suitable layout instead of one partition per key.

This action solves a different issue: “A high-cardinality filter benefits from clustering rather than massive Hive partitions.” Its normal signal is File count and selective scan cost decrease without partition explosion. Here the requested behavior is “The table's data is well laid out but a query still performs full scans because no filter was supplied,” so that signal is not sufficient.

Rewrite the predicate to a sargable equivalent while preserving semantics is appropriate when “A query wraps a partition/filter column in a nonpushdown-compatible function”; it would be verified by Physical plan pushes the predicate and scans fewer files. It does not directly satisfy “The table's data is well laid out but a query still performs full scans because no filter was supplied,” whose relevant acceptance evidence is The plan shows pruning and returned records satisfy the intended subset.

Ensure relevant column statistics and appropriate data layout for pruning belongs to the separate case “A query filters by an indexed or statistically tracked customer/date field,” where Scan metrics show a lower percentage of files read would be expected. That differs from the condition in this stem: “The table's data is well laid out but a query still performs full scans because no filter was supplied.”

That observation is a valid indicator for “A query wraps a partition/filter column in a nonpushdown-compatible function” when using Rewrite the predicate to a sargable equivalent while preserving semantics. The scenario instead tests “The table's data is well laid out but a query still performs full scans because no filter was supplied”; the relevant result is The plan shows pruning and returned records satisfy the intended subset.

Scan metrics show a lower percentage of files read would demonstrate the separate action Ensure relevant column statistics and appropriate data layout for pruning, which addresses “A query filters by an indexed or statistically tracked customer/date field.” It does not verify the requirement “The table's data is well laid out but a query still performs full scans because no filter was supplied” following Add a semantically correct selective WHERE condition to the query.

File count and selective scan cost decrease without partition explosion would demonstrate the separate action Use liquid clustering or suitable layout instead of one partition per key, which addresses “A high-cardinality filter benefits from clustering rather than massive Hive partitions.” It does not verify the requirement “The table's data is well laid out but a query still performs full scans because no filter was supplied” following Add a semantically correct selective WHERE condition to the query.