Free Access to Snowflake.COF-C02.v2025-08-26.q885 with Valid Practice Test (Page 131)

Question 646

What action should be taken if a large number of concurrent queries are queued in a virtual warehouse?

A.Scale-up by resizing the warehouse.
B.Scale-out with a multi-cluster warehouse.
C.Disable auto-suspend on the warehouse.
D.Enable auto-resume on the warehouse.

Correct Answer: B

When a large number of concurrent queries are queued in a virtual warehouse in Snowflake, the most effective action is to scale out using a multi-cluster warehouse. Here's why:
* Understanding the Problem:
* When many queries are queued, it indicates that the computational resources of the warehouse are insufficient to handle the demand.
* Each virtual warehouse in Snowflake provides a fixed amount of compute resources. A single cluster in the warehouse can only process a finite number of concurrent queries efficiently.
* Scale-Out with Multi-Cluster Warehouse:
* Snowflake's multi-cluster warehouse feature allows a warehouse to dynamically add or remove clusters based on the current workload.
* With scale-out, additional clusters are automatically spun up when the query load increases. This helps distribute the workload and reduces queuing without impacting ongoing query performance.
* Once the load decreases, the additional clusters are automatically suspended, optimizing costs.
* Why Not Other Options?:
* A. Scale-up by resizing the warehouse: While scaling up (choosing a larger size, e.g., from X- Small to Medium) increases compute power, it may not efficiently resolve queuing when many concurrent queries need to be processed. It's better suited for improving the performance of individual queries rather than handling more queries simultaneously.
* C. Disable auto-suspend on the warehouse: This ensures the warehouse remains active but does not address the queuing issue. Keeping a warehouse perpetually active can lead to unnecessary costs.
* D. Enable auto-resume on the warehouse: This ensures the warehouse activates when needed but does not directly impact query concurrency.
* Best Practices for Scaling with Multi-Cluster Warehouses:
* Configure the minimum and maximum cluster settings to allow Snowflake to handle varying workloads efficiently.
* Use auto-scaling policies like "economy" or "standard" to control cluster behavior based on cost or performance requirements.
* Steps to Implement Multi-Cluster Warehouse:
* Navigate to the Snowflake console.
* Go to the "Warehouses" tab and select the target warehouse.
* Edit the warehouse settings to set:
* Minimum Clusters: The minimum number of active clusters.
* Maximum Clusters: The upper limit of clusters that Snowflake can scale out to.
* Auto-Resume and Auto-Suspend: Ensure these are enabled for cost efficiency.
References:
* Snowflake Documentation: Scaling a Multi-Cluster Warehouse
* Snowflake Documentation: Query Processing and Queuing
* Snowflake Best Practices for Warehouses

Question 647

What does a Query Profile metric that shows excessive spillage indicate?

A.Improper virtual warehouse sizing
B.Poor query optimization
C.Improper Common Table Expressions (CTEs)
D.Expired temporary tables

Question 648

A user has 10 files in a stage containing new customer dat
a. The ingest operation completes with no errors, using the following command:
COPY INTO my__table FROM @my__stage;
The next day the user adds 10 files to the stage so that now the stage contains a mixture of new customer data and updates to the previous data. The user did not remove the 10 original files.
If the user runs the same copy into command what will happen?

A.All data from all of the files on the stage will be appended to the table
B.Only data about new customers from the new files will be appended to the table
C.The operation will fail with the error uncertain files in stage.
D.All data from only the newly-added files will be appended to the table.

Question 649

True or False: A Virtual Warehouse can be resized while suspended.

A.True
B.False

Question 650

If a Snowflake user decides a table should be clustered, what should be used as the cluster key?

A.The columns that are queried in the select clause.
B.The columns with very high cardinality.
C.The columns with many different values.
D.The columns most actively used in the select filters.

Question 646

Question 647

Question 648

Question 649

Question 650

Download PDF File