Question 350An engineer sees high queueing on a warehouse and enables Query Acceleration Service without considering concurrency. An engineer proposes to investigate multi-cluster scale-out for queue-bound queries before adding acceleration. Which explanation best describes why this is appropriate? Query Acceleration Service can offload eligible portions of a query to additional compute. Query acceleration targets eligible execution work, not warehouse admission contention. Search optimization maintains an access path useful for eligible high-selectivity lookups. Search optimization is not a general replacement for full-table scan processing. ExplanationCorrect Answer B:Query acceleration targets eligible execution work, not warehouse admission contention. This is testable: peak queuing declines after appropriate scaling under controlled parallel requests. Incorrect Answers A:Query Acceleration Service can offload eligible portions of a query to additional compute. For the current requirement, the decisive mechanism is: Query acceleration targets eligible execution work, not warehouse admission contention. The related action is: Investigate multi-cluster scale-out for queue-bound queries before adding acceleration. C:Search optimization maintains an access path useful for eligible high-selectivity lookups. For the current requirement, the decisive mechanism is: Query acceleration targets eligible execution work, not warehouse admission contention. The related action is: Investigate multi-cluster scale-out for queue-bound queries before adding acceleration. D:Search optimization is not a general replacement for full-table scan processing. For the current requirement, the decisive mechanism is: Query acceleration targets eligible execution work, not warehouse admission contention. The related action is: Investigate multi-cluster scale-out for queue-bound queries before adding acceleration.
Question 351A broad range scan across most table rows is nominated for search optimization without profiling. After the team decides to confirm eligible selective predicates and expected benefit before enabling search optimization, which observation most directly validates the result? The controlled equality filter reads less data and meets its latency target after optimization. Eligible test queries show a reduction in duration with documented acceleration use. A sample query with selective predicates demonstrates a measurable benefit. Selected columns show improved lookup latency at acceptable maintenance cost. ExplanationCorrect Answer C:A sample query with selective predicates demonstrates a measurable benefit. This matters because search optimization is not a general replacement for full-table scan processing. Incorrect Answers A:Search optimization maintains an access path useful for eligible high-selectivity lookups. This is not the decisive validation for the stated requirement. The relevant evidence is: A sample query with selective predicates demonstrates a measurable benefit. B:Query Acceleration Service can offload eligible portions of a query to additional compute. This is not the decisive validation for the stated requirement. The relevant evidence is: A sample query with selective predicates demonstrates a measurable benefit. D:Targeted access paths have maintenance cost and predicate-support constraints. This is not the decisive validation for the stated requirement. The relevant evidence is: A sample query with selective predicates demonstrates a measurable benefit.
Question 352A query service stores several roles on its user but uses the same default role for all jobs. If the responsible team does not correct this condition, which outcome is the most relevant remaining risk? The scope error leaves consumers without the required data access. Unexpected table ownership interferes with grants and lifecycle operations. Database-role permissions are never activated in the session. An unrelated scheduled job executes with excessive or insufficient permissions. ExplanationCorrect Answer D:An unrelated scheduled job executes with excessive or insufficient permissions. The direct prevention is to set an explicit primary role per job and review any secondary roles it activates. Assigned roles do not mean every session automatically has the same effective privileges. Incorrect Answers A:Database roles provide reusable database-local privilege bundles through account roles. The immediate omission risk in this case is: An unrelated scheduled job executes with excessive or insufficient permissions. B:New-object ownership and CREATE authorization are based on the session primary role. The immediate omission risk in this case is: An unrelated scheduled job executes with excessive or insufficient permissions. C:Database roles are granted within database scope and cannot themselves become session roles. The immediate omission risk in this case is: An unrelated scheduled job executes with excessive or insufficient permissions.
Question 353A team uses Snowpark for a simple SELECT DISTINCT that its existing SQL view already performs adequately. Which Snowflake configuration or operational action most directly addresses this situation? Compare native SQL with Snowpark code for the measured requirement before adding runtime complexity. Add schema compatibility checks and explicit data quality assertions before release. Push selective filters before expensive joins where semantically safe. Break the workflow into supported tasks with clear dependency and failure checkpoints. ExplanationCorrect Answer A:Snowpark enables supported language processing but not every operation needs additional code. The matching action is to compare native SQL with Snowpark code for the measured requirement before adding runtime complexity. A benchmark meets the target with the simpler maintained SQL implementation. Incorrect Answers B:Source evolution can silently change downstream mappings and expectations. The operative distinction here is: Snowpark enables supported language processing but not every operation needs additional code. The targeted change is: Compare native SQL with Snowpark code for the measured requirement before adding runtime complexity. C:Early reduction of rows limits intermediate join work when optimizer and predicates allow. The operative distinction here is: Snowpark enables supported language processing but not every operation needs additional code. The targeted change is: Compare native SQL with Snowpark code for the measured requirement before adding runtime complexity. D:Task graphs provide ordered execution and per-step operational visibility. The operative distinction here is: Snowpark enables supported language processing but not every operation needs additional code. The targeted change is: Compare native SQL with Snowpark code for the measured requirement before adding runtime complexity.
Question 354A cloud storage stage is configured with a role that can read every bucket in the organization. An engineer proposes to restrict the storage integration's allowed locations and cloud access scope. Which explanation best describes why this is appropriate? Git integration connects version-controlled code assets to authorized Snowflake workflows. Storage integrations couple Snowflake permissions to cloud IAM policies and approved resource paths. The connector provides maintained SQL session and data retrieval behavior. API integrations govern eligible external endpoints and execution authorization. ExplanationCorrect Answer B:Storage integrations couple Snowflake permissions to cloud IAM policies and approved resource paths. This is testable: the stage can read its designated prefix but cannot list a restricted bucket. Incorrect Answers A:Git integration connects version-controlled code assets to authorized Snowflake workflows. For the current requirement, the decisive mechanism is: Storage integrations couple Snowflake permissions to cloud IAM policies and approved resource paths. The related action is: Restrict the storage integration's allowed locations and cloud access scope. C:The connector provides maintained SQL session and data retrieval behavior. For the current requirement, the decisive mechanism is: Storage integrations couple Snowflake permissions to cloud IAM policies and approved resource paths. The related action is: Restrict the storage integration's allowed locations and cloud access scope. D:API integrations govern eligible external endpoints and execution authorization. For the current requirement, the decisive mechanism is: Storage integrations couple Snowflake permissions to cloud IAM policies and approved resource paths. The related action is: Restrict the storage integration's allowed locations and cloud access scope.
Question 355A model owner needs a managed registry for approved model versions and inference lifecycle records. After the team decides to use supported Snowflake ML model-registry capabilities with approved governance, which observation most directly validates the result? The job creates expected derived results without a bulk local extract. Representative questions produce reviewed SQL matching approved measures and filters. A reviewed model version is discoverable and invoked in the approved environment. A sample query returns the expected response shape under the approved role. ExplanationCorrect Answer C:A reviewed model version is discoverable and invoked in the approved environment. This matters because a model registry provides versioned model artifacts and tracked deployment metadata. Incorrect Answers A:Snowpark lets supported language logic execute close to Snowflake-managed data. This is not the decisive validation for the stated requirement. The relevant evidence is: A reviewed model version is discoverable and invoked in the approved environment. B:Cortex Analyst maps questions to SQL using semantic definitions rather than arbitrary table guesses. This is not the decisive validation for the stated requirement. The relevant evidence is: A reviewed model version is discoverable and invoked in the approved environment. D:Cortex AI SQL functions apply hosted AI capabilities to selected content inside Snowflake workflows. This is not the decisive validation for the stated requirement. The relevant evidence is: A reviewed model version is discoverable and invoked in the approved environment.
Question 356Two departments create same-named tables in different schemas and analysts run unqualified queries. If the responsible team does not correct this condition, which outcome is the most relevant remaining risk? The provider exposes the wrong objects because share scope is misunderstood. Table privileges remain unusable because a parent namespace is inaccessible. Identifier generation fails when a context-specific sequence name cannot resolve. Queries silently target an identically named table in another schema. ExplanationCorrect Answer D:Queries silently target an identically named table in another schema. The direct prevention is to specify the intended database and schema in the session or fully qualify the table. Name resolution follows the current database/schema context unless a full object name is supplied. Incorrect Answers A:Shares expose specifically authorized database objects to consumers rather than storing application rows. The immediate omission risk in this case is: Queries silently target an identically named table in another schema. B:Authorization to reach a table requires visibility of its enclosing namespace. The immediate omission risk in this case is: Queries silently target an identically named table in another schema. C:Sequences are schema objects whose names resolve under the current namespace. The immediate omission risk in this case is: Queries silently target an identically named table in another schema.
Question 357A reporting team needs a virtual representation of data without storing a new full copy. Which Snowflake configuration or operational action most directly addresses this situation? Create a standard view over the governed source objects. Explain standard-view evaluation against source tables at query time. Consider a secure view with explicit access policies rather than an unrestricted standard view. Trace dependencies and analyze query plans before choosing another view layer. ExplanationCorrect Answer A:A standard view saves a query definition rather than materializing its full result table. The matching action is to create a standard view over the governed source objects. Reading the view returns current authorized source rows without a separate reload. Incorrect Answers B:Standard views evaluate their definition using current authorized data. The operative distinction here is: A standard view saves a query definition rather than materializing its full result table. The targeted change is: Create a standard view over the governed source objects. C:Secure views restrict certain optimizations and metadata exposure for protected sharing patterns. The operative distinction here is: A standard view saves a query definition rather than materializing its full result table. The targeted change is: Create a standard view over the governed source objects. D:A view is a logical SQL abstraction and excessive nesting can complicate optimization and diagnosis. The operative distinction here is: A standard view saves a query definition rather than materializing its full result table. The targeted change is: Create a standard view over the governed source objects.
Question 358A team changes the source of a revenue metric but cannot trace which downstream dashboard it affects. An engineer proposes to inspect available data lineage and object dependencies before releasing the change. Which explanation best describes why this is appropriate? Consumers and derived objects can depend on the same source across schemas. Lineage links transformations and consuming objects to upstream data assets. Tags create inspectable metadata for classification and policy workflows. Tag metadata can drive policy assignment at scale when configured deliberately. ExplanationCorrect Answer B:Lineage links transformations and consuming objects to upstream data assets. This is testable: the analysis identifies the affected materialization and reporting consumers. Incorrect Answers A:Consumers and derived objects can depend on the same source across schemas. For the current requirement, the decisive mechanism is: Lineage links transformations and consuming objects to upstream data assets. The related action is: Inspect available data lineage and object dependencies before releasing the change. C:Tags create inspectable metadata for classification and policy workflows. For the current requirement, the decisive mechanism is: Lineage links transformations and consuming objects to upstream data assets. The related action is: Inspect available data lineage and object dependencies before releasing the change. D:Tag metadata can drive policy assignment at scale when configured deliberately. For the current requirement, the decisive mechanism is: Lineage links transformations and consuming objects to upstream data assets. The related action is: Inspect available data lineage and object dependencies before releasing the change.
Question 359A new Snowpipe definition exists but cloud object notifications never reach it. After the team decides to verify the external stage integration and notification channel end to end, which observation most directly validates the result? The corrected test file loads successfully with its prior error resolved. Service billing and pipe history identify ingestion without a continuously running BI warehouse. Uploading one test object produces an observable pipe event and target rows. New test files appear in the table within the documented operational target. ExplanationCorrect Answer C:Uploading one test object produces an observable pipe event and target rows. This matters because auto-ingest depends on a supported event path between object storage and the pipe. Incorrect Answers A:Snowpipe uses configured COPY logic and exposes ingestion error evidence. This is not the decisive validation for the stated requirement. The relevant evidence is: Uploading one test object produces an observable pipe event and target rows. B:Snowpipe runs its supported ingestion service separately from user-managed warehouse COPY runs. This is not the decisive validation for the stated requirement. The relevant evidence is: Uploading one test object produces an observable pipe event and target rows. D:Snowpipe continuously loads eligible new staged files through managed ingest compute. This is not the decisive validation for the stated requirement. The relevant evidence is: Uploading one test object produces an observable pipe event and target rows.
Question 360A pipeline loads JSON strings into VARCHAR and later tries to access nested keys as VARIANT values. If the responsible team does not correct this condition, which outcome is the most relevant remaining risk? Lexical comparison produces incorrect monetary totals. Nested items remain inaccessible to relational aggregation. Unexpected row explosions inflate metrics and compute cost. Ad hoc string parsing yields brittle or missing nested values. ExplanationCorrect Answer D:Ad hoc string parsing yields brittle or missing nested values. The direct prevention is to parse the JSON into a VARIANT representation before path queries. VARIANT preserves semi-structured types and supports documented element-path access. Incorrect Answers A:Typed conversion clarifies semantics for numeric predicates and calculations. The immediate omission risk in this case is: Ad hoc string parsing yields brittle or missing nested values. B:FLATTEN expands array or object elements to relational rows for downstream SQL. The immediate omission risk in this case is: Ad hoc string parsing yields brittle or missing nested values. C:Expansion multiplies source rows by the selected collection's elements. The immediate omission risk in this case is: Ad hoc string parsing yields brittle or missing nested values.