Chapter 8: Performance Flashcards
Which of the following is true when a virtual warehouse is scaled up to a larger size? (Choose all that apply.)
Charging for the new size does not begin until all the new nodes in the larger virtual warehouse have been provisioned.
Charging for the new size begins immediately.
Any queries that were already running on the virtual warehouse do not benefit from the new size.
Only new queries can take advantage of the increased cluster size.
A virtual warehouse cannot be scaled up if there are any active queries on the virtual warehouse.
Charging for the new size does not begin until all the new nodes in the larger virtual warehouse have been provisioned.
Any queries that were already running on the virtual warehouse do not benefit from the new size.
Only new queries can take advantage of the increased cluster size.
Which of the following is true when a virtual warehouse is scaled down to a smaller size? (Choose all that apply.)
When a virtual warehouse is scaled down, nodes are withdrawn from the compute cluster.
When a virtual warehouse is scaled down, nodes are removed only when there are no active queries on the nodes that are to be removed.
The decrease in size does not impact any queries that were already running on the virtual warehouse.
When a virtual warehouse is scaled down, nodes are withdrawn from the compute cluster.
When a virtual warehouse is scaled down, nodes are removed only when there are no active queries on the nodes that are to be removed.
The decrease in size does not impact any queries that were already running on the virtual warehouse.
Scaling up a virtual warehouse to a larger size is suitable for which of the following scenarios?
There are a large number of concurrent users on the system.
There are a large number of concurrent queries on the system.
Complex queries are running on the system and are required to finish faster.
Complex queries are running on the system and are required to finish faster.
Which of the following statements describe scaling out in Snowflake? (Choose all that apply.)
Scaling out is achieved by using multi-cluster virtual warehouses.
Scaling out is achieved by increasing or decreasing a virtual warehouse size.
Scaling out can help reduce query queuing.
Scaling out is achieved by using multi-cluster virtual warehouses.
Scaling out can help reduce query queuing.
What is the minimum Snowflake edition that supports multi-cluster virtual warehouses?
Standard
Enterprise
Business Critical
Virtual Private Snowflake
Enterprise
Which of the following scaling policies aims to minimize query queuing?
Standard
Economy
Standard
Consider the following configuration and select the mode of operation for this multi-cluster virtual warehouse.
Minimum cluster count = 2
Maximum cluster count = 5
Size = Large
Maximized
Autoscale
Maximum cluster count = 5
Which of the following statements are true regarding a multi-cluster virtual warehouse in autoscale mode?
Additional warehouses are started up when Snowflake detects that queries are starting to queue.
Warehouses are progressively shut down when the query demand reduces.
Multi-cluster virtual warehouses support all the properties and actions as regular virtual warehouses. They can be suspended, resumed, and reconfigured to be of a different size.
Once a multi-cluster virtual warehouse is created, its configuration cannot be changed. It cannot be suspended or resumed.
Additional warehouses are started up when Snowflake detects that queries are starting to queue.
Warehouses are progressively shut down when the query demand reduces.
Multi-cluster virtual warehouses support all the properties and actions as regular virtual warehouses. They can be suspended, resumed, and reconfigured to be of a different size.
Which of the following conditions must be met for Snowflake to reuse the query result cache for a query? (Choose all that apply.)
The query does not use runtime functions.
The underlying data contributing to the query results has not changed.
A new query matches an old query.
Reclustering or consolidation has not changed the table micro-partitions.
Neither user-defined nor external functions are used in the query.
The query does not use runtime functions.
The underlying data contributing to the query results has not changed.
A new query matches an old query.
Reclustering or consolidation has not changed the table micro-partitions.
Neither user-defined nor external functions are used in the query.
Which of the following are true regarding the query result cache? (Choose all that apply.)
The query result cache is used if an identical query is run within a 24-hour period of the original query (that produced the query result cache).
The maximum number of days for which a query result cache may be retained is 31 days.
The query result cache is stored in virtual warehouse SSD storage.
The query result cache is stored in the cloud services layer.
The query result cache is purged if it is not reused within 24 hours.
The query result cache is used if an identical query is run within a 24-hour period of the original query (that produced the query result cache).
The maximum number of days for which a query result cache may be retained is 31 days
The query result cache is purged if it is not reused within 24 hours.
The query result cache is stored in the cloud services layer.
Which of the following are examples of caches in Snowflake? (Choose all that apply.)
Query result cache
Cloud platform cache
Virtual warehouse cache
Metadata cache
High-speed cache
Query result cache
Virtual warehouse cache
Metadata cache
To process SQL queries, a virtual warehouse must generally be running. Which types of queries can produce results without the need for a running virtual warehouse? (Choose all that apply.)
Queries that have previously been run and have produced a result cache
Queries such as count of rows, minimum and maximum for a column
Queries that produce less than 100 MB of a result set
Queries that use only one table
Queries that have previously been run and have produced a result cache
Queries such as count of rows, minimum and maximum for a column
Which of the following caches are stored in the cloud services layer? (Choose all that apply.)
Query result cache
Metadata cache
Virtual warehouse cache
Browser cache
SnowSQL cache
Query result cache
Metadata cache
How does clustering keys improve query performance?
By precalculating query results and physically storing them
By compressing data in columns
By distributing data in micro-partitions such that a more optimized partition pruning can occur during query execution
By distributing the data over multiple compute clusters so that each computer cluster has a subset of data to process
By distributing data in micro-partitions such that a more optimized partition pruning can occur during query execution
Which of the following are true regarding materialized views? (Choose all that apply.)
A materialized view automatically updates when the underlying table’s data changes.
Materialized views are used to improve query performance. They precompute query results and store the results physically.
A materialized view can provide precomputed results that can fulfill some queries much faster.
Materialized views are automatically maintained by a Snowflake service running in the background, transparent to the users.
A materialized view automatically updates when the underlying table’s data changes.
Materialized views are used to improve query performance. They precompute query results and store the results physically.
A materialized view can provide precomputed results that can fulfill some queries much faster.
Materialized views are automatically maintained by a Snowflake service running in the background, transparent to the users.
