Free Access to Snowflake.GES-C01.v2026-03-27.q130 with Valid Practice Test (Page 19)

Question 86

A financial institution uses Snowflake Cortex Analyst with strict role-based access control (RBAC) on their Snowflake-hosted LLMs. The security team has granted specific 'CORTEX-MODEL-ROLE application roles to different analyst teams, ensuring they only access approved models. A new requirement arises to enable Azure OpenAI GPT models for Cortex Analyst to leverage a specific feature. An administrator proceeds to execute:

Which of the following statements accurately describe the implications of this change?

A.Option A
B.Option B
C.Option C
D.Option D
E.Option E

Question 87

A data scientist, 'Dl DEV', has been granted the

Despite these grants, 'DI_DEV' still receives a 'permission denied' error when attempting to ''prepare'' the Document AI model build in Snowsight. Which 'single missing privilege' is the most likely direct cause of this specific error for Document AI model build preparation?

A.
B.
C.
D.
E.

Question 88

A business team using a Snowflake Cortex Analyst-powered chatbot reports that follow-up questions in multi-turn conversations are sometimes slow to process, impacting user experience. The development team wants to optimize for responsiveness while maintaining accuracy in SQL generation. Which of the following strategies directly addresses latency in multi-turn conversations within Cortex Analyst, considering its underlying mechanisms?

A.Increase the warehouse size used for Cortex Analyst queries to 'Large' to accelerate LLM inference.
B.Implement an explicit LLM summarization agent within the semantic model to condense conversation history before it's passed to subsequent LLM calls.
C.Configure the semantic model to reset the conversation context after every three turns to limit token count.
D.Switch the underlying text-to-SQL LLM to a smaller model, such as
E.Rely on

Question 89

A development team plans to utilize Snowpark Container Services (SPCS) for deploying a variety of AI/ML workloads, including custom LLMs and GPU-accelerated model training jobs. They are in the process of creating a compute pool and need to select the appropriate instance families and configurations. Which of the following statements about 'CREATE COMPUTE POOL' in SPCS are accurate?

A.To support GPU-accelerated LLM inference and training, the 'INSTANCE_FAMILY' must be selected from a type starting with 'GPU' (e.g.,
B.The 'MIN NODES' and 'MAX NODES parameters define the scaling range for the compute pool, and Snowflake automatically scales the pool within this range based on workload demand.
C.Setting 'AUTO RESUME = TRUE ensures that the compute pool automatically starts when a service or job is submitted to it, rather than requiring manual resumption.
D.For cost optimization, 'AUTO SUSPEND SECS = 0' should be used to prevent automatic suspension of the compute pool, as suspension and resumption incur minimum billing durations.
E.Snowpark-optimized warehouses are the recommended compute pool type for all large-scale ML training workloads within SPCS due to their enhanced memory limits and CPU architectures.

Correct Answer: A,C

Option A is correct. GPU-accelerated workloads, such as LLM inference and model training, require instance families specifically designed with GPUs. The documentation lists instance family names starting with 'GPU' for this purpose, such as or 'GPU_GCP NV L4 Option B is incorrect. While 'MIN NODES and 'MAX NODES define the range, the size of compute clusters in Snowpark Container Services does ''not'' auto-scale dynamically based on workload demand. Users must manually alter the number of instances at runtime using commands like 'ALTER SERVICE MIN INSTANCES = s. Snowflake does handle load balancing across instances within the configured node counts. Option C is correct. The 'AUTO_RESUME = TRUE parameter, when specified during compute pool creation, enables the pool to automatically resume operation when a service or job is submitted, removing the need for explicit SALTER COMPUTE POOL RESUME commands. option D is incorrect. Setting = prevents the compute pool from automatically suspending, meaning it will continue to consume credits even when idle. This would generally lead to higher costs, not cost optimization, unless the pool is constantly active. The default is 3600 seconds (1 hour). SPCS Compute Nodes have a minimum charge of five minutes when started or resumed, making intelligent use of auto-suspend important for cost management. Option E is incorrect. Snowpark-optimized warehouses are a type of 'virtual warehouse' and are recommended for Snowpark workloads with large memory requirements or specific CPU architecture, typically for single-node ML training workloads 'within a warehouse'. SPCS compute pools, however, provide their own dedicated instance families (CPU, HighMemory, GPU) for containerized workloads, abstracting the underlying infrastructure and supporting distributed GPU clusters directly within SPCS, not Snowpark-optimized warehouses as a 'compute pool type' for SPCS.

Question 90

A development team is implementing a document retrieval system in Snowflake. They plan to store document embeddings and use VECTOR_L2_DISTANCE to find the most relevant documents for a given query embedding. Considering Snowflake's capabilities, which of the following statements are true regarding the use of vector types and VECTOR_L2_DISTANCE
? (Select all that apply)

A.Document embeddings, which are typically float arrays, can be stored in a
B.O When defining a table column for 1024-dimensional float embeddings, the SQL type specification
C.Using the Snowpark Python library, developers can directly invoke
D.VECTOR
E.To prevent issues with direct vector comparisons, explicitly using

Question 86

Question 87

Question 88

Question 89

Question 90

Download PDF File