Question 370A dashboard always filters by highly selective customer key while data is organized only by event date. An engineer proposes to compare search optimization and clustering using real predicate frequencies and costs. Which explanation best describes why this is appropriate? Eligible materialized views store maintained results that can avoid repeated full recomputation. Different optimization techniques suit point lookups versus broad range pruning. Small tables often cannot amortize clustering maintenance against query savings. A standard view stores its SQL definition; materialized views maintain results separately. ExplanationCorrect Answer B:Different optimization techniques suit point lookups versus broad range pruning. This is testable: the selected technique meets customer-key latency with justified maintenance charges. Incorrect Answers A:Eligible materialized views store maintained results that can avoid repeated full recomputation. For the current requirement, the decisive mechanism is: Different optimization techniques suit point lookups versus broad range pruning. The related action is: Compare search optimization and clustering using real predicate frequencies and costs. C:Small tables often cannot amortize clustering maintenance against query savings. For the current requirement, the decisive mechanism is: Different optimization techniques suit point lookups versus broad range pruning. The related action is: Compare search optimization and clustering using real predicate frequencies and costs. D:A standard view stores its SQL definition; materialized views maintain results separately. For the current requirement, the decisive mechanism is: Different optimization techniques suit point lookups versus broad range pruning. The related action is: Compare search optimization and clustering using real predicate frequencies and costs.
Question 371An engineer changes SQL repeatedly without saving the prior query IDs and timing conditions. After the team decides to record query IDs, warehouse configuration, cache state and result validity for each test, which observation most directly validates the result? The target statement is associated with its user, query ID and time range. The requested historical query appears in the permitted account-usage view. The benchmark links each measured run to a reproducible configuration. Cost totals reconcile to billing and per-team workload estimates. ExplanationCorrect Answer C:The benchmark links each measured run to a reproducible configuration. This matters because performance comparisons are meaningful only with comparable inputs and execution environments. Incorrect Answers A:Snowflake execution metadata ties statements to recorded session actors. This is not the decisive validation for the stated requirement. The relevant evidence is: The benchmark links each measured run to a reproducible configuration. B:Snowsight's recent-activity window differs from longer-term account usage availability. This is not the decisive validation for the stated requirement. The relevant evidence is: The benchmark links each measured run to a reproducible configuration. D:Shared compute costs require workload-aware allocation rather than an equal headcount assumption. This is not the decisive validation for the stated requirement. The relevant evidence is: The benchmark links each measured run to a reproducible configuration.
Question 372A reloadable integration dataset needs low-cost storage but no Fail-safe recovery guarantee. If the responsible team does not correct this condition, which outcome is the most relevant remaining risk? A proprietary-only table design blocks planned cross-engine access. Sensitive scratch data survives longer than the intended working session. A manually rebuilt table becomes stale between runs. The design either pays for unneeded protection or promises unsupported recovery. ExplanationCorrect Answer D:The design either pays for unneeded protection or promises unsupported recovery. The direct prevention is to evaluate a transient table with explicitly acceptable recovery tradeoffs. Transient tables omit Fail-safe and have more limited recovery than permanent tables. Incorrect Answers A:Iceberg provides an open table format with governed catalog integration across engines. The immediate omission risk in this case is: The design either pays for unneeded protection or promises unsupported recovery. B:Temporary tables are visible only within their session and are removed when the session ends. The immediate omission risk in this case is: The design either pays for unneeded protection or promises unsupported recovery. C:Dynamic tables maintain query-defined results through managed refresh scheduling. The immediate omission risk in this case is: The design either pays for unneeded protection or promises unsupported recovery.
Question 373A scheduled transform has no deterministic way to detect data drift when a source schema adds a new field. Which Snowflake configuration or operational action most directly addresses this situation? Add schema compatibility checks and explicit data quality assertions before release. Choose appropriate ephemeral, transient or dynamic outputs based on recovery and freshness needs. Adopt a supported incremental transformation using change capture and keyed MERGE. Apply appropriate authorization and filtering at source before publishing derived results. ExplanationCorrect Answer A:Source evolution can silently change downstream mappings and expectations. The matching action is to add schema compatibility checks and explicit data quality assertions before release. A sample unexpected column triggers a controlled validation alert. Incorrect Answers B:Table lifecycle choices govern persistence, recovery overhead and maintenance behavior. The operative distinction here is: Source evolution can silently change downstream mappings and expectations. The targeted change is: Add schema compatibility checks and explicit data quality assertions before release. C:Incremental processing can apply changed records while retaining stable target data. The operative distinction here is: Source evolution can silently change downstream mappings and expectations. The targeted change is: Add schema compatibility checks and explicit data quality assertions before release. D:Downstream transformations cannot replace source-access governance for sensitive data. The operative distinction here is: Source evolution can silently change downstream mappings and expectations. The targeted change is: Add schema compatibility checks and explicit data quality assertions before release.
Question 374A staging table should disappear when a data engineer's session ends. An engineer proposes to create a temporary table rather than a permanent table for session-scoped scratch data. Which explanation best describes why this is appropriate? Dynamic tables maintain query-defined results through managed refresh scheduling. Temporary tables are visible only within their session and are removed when the session ends. Iceberg provides an open table format with governed catalog integration across engines. Permanent tables can retain historical versions for supported recovery operations. ExplanationCorrect Answer B:Temporary tables are visible only within their session and are removed when the session ends. This is testable: a fresh independent session cannot query the scratch table after the creator exits. Incorrect Answers A:Dynamic tables maintain query-defined results through managed refresh scheduling. For the current requirement, the decisive mechanism is: Temporary tables are visible only within their session and are removed when the session ends. The related action is: Create a temporary table rather than a permanent table for session-scoped scratch data. C:Iceberg provides an open table format with governed catalog integration across engines. For the current requirement, the decisive mechanism is: Temporary tables are visible only within their session and are removed when the session ends. The related action is: Create a temporary table rather than a permanent table for session-scoped scratch data. D:Permanent tables can retain historical versions for supported recovery operations. For the current requirement, the decisive mechanism is: Temporary tables are visible only within their session and are removed when the session ends. The related action is: Create a temporary table rather than a permanent table for session-scoped scratch data.
Question 375A production table is accidentally dropped and an administrator needs it restored within the configured Time Travel window. After the team decides to use UNDROP TABLE under the eligible table and retention conditions, which observation most directly validates the result? The runbook documents the supported recovery boundary and escalation process. A cloned test table is created quickly and its later updates do not mutate production. The restored table returns the expected row count and authorized query results. A controlled comparison returns yesterday's and today's distinct approved metrics. ExplanationCorrect Answer C:The restored table returns the expected row count and authorized query results. This matters because time Travel supports restoring eligible historical table states before the retention window expires. Incorrect Answers A:Fail-safe is a last-resort recovery mechanism, not a normal customer SQL time-travel query path. This is not the decisive validation for the stated requirement. The relevant evidence is: The restored table returns the expected row count and authorized query results. B:Clones initially reference shared existing micro-partitions and diverge as data changes. This is not the decisive validation for the stated requirement. The relevant evidence is: The restored table returns the expected row count and authorized query results. D:Historical query syntax accesses retained earlier data versions within the supported window. This is not the decisive validation for the stated requirement. The relevant evidence is: The restored table returns the expected row count and authorized query results.
Question 376A provider edits a published listing description without verifying that its data product still supports the promised schema. If the responsible team does not correct this condition, which outcome is the most relevant remaining risk? Customers receive unmanaged scripts with no reviewed permission boundary. A sensitive product is exposed to an unintended general audience. Customers expect unrestricted source access and encounter blocked queries. New subscribers receive misleading schema promises and broken integrations. ExplanationCorrect Answer D:New subscribers receive misleading schema promises and broken integrations. The direct prevention is to validate the delivered shared objects and consumer queries before updating the listing claim. Listing documentation should match the live governed data product contract. Incorrect Answers A:Native Apps distribute supported application capabilities under the consumer's account controls. The immediate omission risk in this case is: New subscribers receive misleading schema promises and broken integrations. B:Private listings restrict audience while retaining listing-based discovery for invited accounts. The immediate omission risk in this case is: New subscribers receive misleading schema promises and broken integrations. C:Listing discovery and underlying data authorization are distinct steps. The immediate omission risk in this case is: New subscribers receive misleading schema promises and broken integrations.
Question 377A data provider wants discovery by a broad customer audience rather than sending a direct share to each known account. Which Snowflake configuration or operational action most directly addresses this situation? Evaluate a public Snowflake Marketplace listing with reviewed description and access requirements. Evaluate a Snowflake Native App package and its permission model. Validate the delivered shared objects and consumer queries before updating the listing claim. Review the listing's actual provider grants, terms and required approval flow. ExplanationCorrect Answer A:Marketplace listings provide a catalog distribution path for eligible shared data products. The matching action is to evaluate a public Snowflake Marketplace listing with reviewed description and access requirements. A test consumer discovers the listing and requests or obtains authorized access. Incorrect Answers B:Native Apps distribute supported application capabilities under the consumer's account controls. The operative distinction here is: Marketplace listings provide a catalog distribution path for eligible shared data products. The targeted change is: Evaluate a public Snowflake Marketplace listing with reviewed description and access requirements. C:Listing documentation should match the live governed data product contract. The operative distinction here is: Marketplace listings provide a catalog distribution path for eligible shared data products. The targeted change is: Evaluate a public Snowflake Marketplace listing with reviewed description and access requirements. D:Listing discovery and underlying data authorization are distinct steps. The operative distinction here is: Marketplace listings provide a catalog distribution path for eligible shared data products. The targeted change is: Evaluate a public Snowflake Marketplace listing with reviewed description and access requirements.
Question 378A finance export produces one giant CSV that downstream workers cannot process concurrently. An engineer proposes to choose suitable unloading file-size and partitioning settings for the receiver. Which explanation best describes why this is appropriate? Storage integrations avoid copying raw cloud keys into each pipeline configuration. COPY INTO location can distribute results across files for scalable consumption. Unloading writes query results to a stage-backed destination rather than exporting them via SELECT alone. Unloading materializes the results of the authorized query, so access scope must be correct first. ExplanationCorrect Answer B:COPY INTO location can distribute results across files for scalable consumption. This is testable: the export produces expected file sizes and complete total record counts. Incorrect Answers A:Storage integrations avoid copying raw cloud keys into each pipeline configuration. For the current requirement, the decisive mechanism is: COPY INTO location can distribute results across files for scalable consumption. The related action is: Choose suitable unloading file-size and partitioning settings for the receiver. C:Unloading writes query results to a stage-backed destination rather than exporting them via SELECT alone. For the current requirement, the decisive mechanism is: COPY INTO location can distribute results across files for scalable consumption. The related action is: Choose suitable unloading file-size and partitioning settings for the receiver. D:Unloading materializes the results of the authorized query, so access scope must be correct first. For the current requirement, the decisive mechanism is: COPY INTO location can distribute results across files for scalable consumption. The related action is: Choose suitable unloading file-size and partitioning settings for the receiver.
Question 379A multi-cluster warehouse holds five clusters all night despite a configured low minimum. After the team decides to review actual scaling mode, running workload and suspend behavior before changing caps, which observation most directly validates the result? Warehouse Load History shows fewer extra clusters while waits stay within policy. The warehouse adds a second cluster during a controlled concurrency test. Cluster events and warehouse activity explain the active count and subsequent scale-in. Peak dashboard requests show the intended queue-latency reduction. ExplanationCorrect Answer C:Cluster events and warehouse activity explain the active count and subsequent scale-in. This matters because cluster scale-in depends on policy and sustained workload demand, not a static promise. Incorrect Answers A:Economy waits longer before provisioning another cluster to favor utilization. This is not the decisive validation for the stated requirement. The relevant evidence is: Cluster events and warehouse activity explain the active count and subsequent scale-in. B:The maximum cluster count is a hard upper bound on simultaneous warehouse clusters. This is not the decisive validation for the stated requirement. The relevant evidence is: Cluster events and warehouse activity explain the active count and subsequent scale-in. D:Standard responds more aggressively to workload queue pressure. This is not the decisive validation for the stated requirement. The relevant evidence is: Cluster events and warehouse activity explain the active count and subsequent scale-in.
Question 380A report selects one customer but Query Profile shows almost every micro-partition scanned. If the responsible team does not correct this condition, which outcome is the most relevant remaining risk? The join remains I/O-bound while the team tunes unrelated filters. Downstream aggregates and warehouse use both grow from spurious rows. An untargeted compute increase misses the expensive plan step. The query keeps reading irrelevant storage pages despite returning few rows. ExplanationCorrect Answer D:The query keeps reading irrelevant storage pages despite returning few rows. The direct prevention is to inspect predicate selectivity and pruning evidence before changing warehouse size. Poor pruning can dominate scans even when result sets are small. Incorrect Answers A:Large intermediate hash tables can exceed available memory and spill to slower storage. The immediate omission risk in this case is: The query keeps reading irrelevant storage pages despite returning few rows. B:Missing predicates multiply matching rows and inflate intermediate result size. The immediate omission risk in this case is: The query keeps reading irrelevant storage pages despite returning few rows. C:Operator-level execution statistics isolate scan, join, aggregation and spill costs. The immediate omission risk in this case is: The query keeps reading irrelevant storage pages despite returning few rows.