ARA-C01 Snowflake Architecture Practice Test 6 — 15 Questions

Question 1

Which feature of Snowflake's architecture allows for the scaling of resources to handle high-concurrency BI dashboard requests without affecting the performance of heavy ETL workloads?

Accepted Answer

Independent Virtual Warehouses. Snowflake's multi-cluster warehouse architecture enables the separation of workloads. By assigning distinct Virtual Warehouses to different user groups or applications, architects ensure that one workload cannot starve another for resources. This prevents contention and allows for granular, independent scaling based on the specific needs of each workload, which is a core architectural benefit for enterprise-level data environments managing diverse user requirements.

Answer

Warehouse Resizing

Answer

Data Sharing

Answer

Automatic Query Clustering

Question 2

What happens when a Virtual Warehouse is 'resized' to a larger size while queries are currently running on it?

Accepted Answer

Running queries continue on the original resources; new queries use the new size.. Resizing a warehouse is a dynamic operation. When you increase the size, Snowflake adds new compute nodes to the existing cluster. The currently running queries continue to run on the old nodes, while all new queries submitted after the resize operation begin using the new, larger capacity. This allows for seamless scaling without interrupting or failing active queries, which is critical for high-availability production environments requiring zero downtime.

Answer

All running queries are terminated and restarted.

Answer

The warehouse is paused until all current queries finish.

Answer

The warehouse creates a new temporary warehouse.

Question 3

A global e-commerce company stores order events in a Snowflake table that ingests roughly 40 million rows per day. Analysts report that point-lookup queries filtering on ORDER_ID are fast, but range queries that filter on ORDER_DATE and aggregate by REGION scan almost every partition and take several minutes. The architect must reduce bytes scanned for the date-range workload while minimizing rebuild cost, and the table already has a natural daily ingestion cadence. Which action best satisfies the requirement?

Accepted Answer

Define a clustering key on (ORDER_DATE, REGION) and enable Automatic Clustering on the table.. The date-range aggregation reads almost every micro-partition, which is a pruning problem rather than a compute or lookup problem. Establishing a clustering key with ORDER_DATE first and REGION second aligns partition boundaries with the dominant filter, and because ingestion is naturally ordered by date, Automatic Clustering can incrementally maintain the layout instead of rebuilding from scratch. That combination directly lowers bytes scanned for the range workload while keeping maintenance cost proportionate to new data.

Answer

Resize the virtual warehouse to a larger multi-cluster size so more compute is available for the scan.

Answer

Create a materialized view that selects ORDER_DATE, REGION, and the aggregate measures, then query the view instead of the base table.

Answer

Enable a search optimization service on the ORDER_DATE column to speed up the range filter.

Question 4

A Snowflake architect is designing a solution where multiple independent compute clusters (virtual warehouses) must access the same underlying data without needing to copy it, and each workload must be isolated to avoid resource contention. Which Snowflake architecture feature directly enables this?

Accepted Answer

Multi-cluster shared data architecture. Snowflake's architecture separates storage and compute, allowing multiple virtual warehouses to access the same central data repository without duplication. This multi-cluster shared data design provides compute isolation and elastic scalability while ensuring all warehouses see consistent data. Replication and cloning create copies, and auto-scaling manages resources within a single warehouse, so they do not meet the requirement of shared data across independent compute clusters.

Answer

Database replication

Answer

Zero-copy cloning

Answer

Virtual warehouse auto-scaling

Question 5

An architect is configuring a Snowflake account to allow users to authenticate via an external OAuth 2.0 identity provider. The security team requires that the OAuth token includes specific custom claims that map to Snowflake roles. Which Snowflake feature should the architect use to enforce role mapping based on these claims?

Accepted Answer

External OAuth integration with a security integration object. External OAuth integration in Snowflake is designed to accept tokens from an external OAuth 2.0 authorization server. By configuring a security integration with the appropriate issuer, audience, and claim mappings, Snowflake can extract custom claims from the token and map them to roles. This allows the identity provider to control role assignments dynamically. Other authentication methods like SAML, key pair, or managed identity do not support OAuth token claims for role mapping.

Answer

Key pair authentication with RSA public/private keys

Answer

SAML 2.0 federation with a security integration

Answer

Snowflake managed identity with Azure AD

Question 6

An architect is evaluating Snowflake's Virtual Warehouse architecture for a new workload that requires continuous, high-concurrency query execution with minimal latency. The workload consists of many small, short-running queries issued by hundreds of users simultaneously. Which configuration should the architect recommend to best meet these requirements?

Accepted Answer

Deploy a single multi-cluster warehouse with a minimum of 3 clusters and a maximum of 10 clusters, using STANDARD scaling policy.. High-concurrency, short-running queries benefit from a multi-cluster warehouse with STANDARD scaling, which aggressively adds clusters to reduce queuing. A minimum of 3 clusters handles baseline concurrency, and a maximum of 10 allows scaling for peaks. ECONOMY scaling is more conservative and may introduce latency, while single-cluster or per-user warehouses do not scale as efficiently.

Answer

Deploy a single large warehouse with a maximum cluster count of 1 and enable auto-suspend after 60 seconds.

Answer

Deploy a single multi-cluster warehouse with ECONOMY scaling policy and a maximum of 2 clusters.

Answer

Deploy multiple independent warehouses, each sized X-Small, and assign each user to a dedicated warehouse.

Question 7

A Snowflake architect is explaining the layers of Snowflake's architecture to a new team. Which layer is responsible for managing metadata, optimizing queries, and providing transaction management?

Accepted Answer

Cloud Services layer. Snowflake's architecture consists of three layers: storage, compute, and Cloud Services. The Cloud Services layer is responsible for metadata management, query optimization, transaction management, and access control. It coordinates all activities and ensures ACID compliance. The compute layer executes queries, and the storage layer stores data. Thus, the Cloud Services layer is the correct answer.

Answer

Compute layer

Answer

Database layer

Answer

Storage layer

Question 8

An architect is designing a data pipeline that ingests JSON data into a Snowflake table with a VARIANT column. The pipeline uses Snowpipe for continuous ingestion. The architect notices that some JSON records contain nested arrays and occasionally exceed 16 MB in size. Which statement accurately describes a limitation or behavior the architect must consider?

Accepted Answer

The VARIANT data type can store up to 16 MB of uncompressed data per value, and exceeding this limit causes the load to fail with an error.. VARIANT columns in Snowflake have a strict 16 MB uncompressed size limit per value. When ingesting JSON via Snowpipe, if any single record exceeds this, the load fails. Architects must preprocess data to split or truncate large records. Compression does not change the uncompressed size limit, and Snowpipe does not split records automatically.

Answer

The VARIANT data type has no size limit, but Snowpipe enforces a 16 MB limit per file.

Answer

Snowpipe compresses JSON data, so the 16 MB limit applies to the compressed size, allowing larger uncompressed records.

Answer

Snowpipe automatically splits large JSON records larger than 16 MB into multiple VARIANT rows to comply with the size limit.

Question 9

An architect is designing a secure data sharing strategy using Snowflake's Data Sharing feature. The goal is to share a subset of data with a partner account while ensuring that the partner cannot access underlying micro-partitions directly and that the shared data is always up-to-date. Which two statements are true regarding Snowflake Data Sharing? (Choose two.)

Accepted Answer

Data sharing provides live, read-only access to the shared data without copying it, ensuring the partner always sees the latest changes.. Snowflake Data Sharing provides live, read-only access to data without copying, and providers use share objects to grant granular privileges. Consumers cannot modify shared data or create micro-partitions. Cross-region sharing is possible with replication, so the same-region requirement is not absolute.

Answer

The consumer can modify the shared data by inserting new rows into the shared tables.

Answer

Data sharing requires the consumer to have a Snowflake account in the same cloud region as the provider.

Answer

The consumer can create their own micro-partitions on the shared data to improve query performance.

Question 10

An architect is explaining Snowflake's architecture to a team new to the platform. The team wants to understand which component is responsible for storing metadata about tables, columns, and micro-partitions to enable efficient query pruning. Which Snowflake layer provides this functionality?

Accepted Answer

Cloud services layer. The cloud services layer is responsible for metadata management, including table definitions, column statistics, and micro-partition metadata used for pruning. It is a key component that enables Snowflake's performance and ease of use, distinct from storage and compute layers.

Answer

Database layer

Answer

Compute layer

Answer

Storage layer

Question 11

An architect is configuring a Snowflake account to support a workload that requires reading data from an external Apache Iceberg table stored in an external cloud storage location. The architect wants to ensure that queries against the Iceberg table can leverage Snowflake's compute and caching. Which Snowflake feature should the architect use?

Accepted Answer

Iceberg tables. Snowflake Iceberg tables enable native integration with Apache Iceberg, allowing Snowflake to query and write to external Iceberg data using its compute and caching. This feature supports ACID transactions and schema evolution, making it ideal for lakehouse architectures. External tables and stages do not provide the same level of integration.

Answer

Secure views

Answer

External tables

Answer

External stages

Question 12

A data architect is designing a multi-cluster warehouse for a retail company's peak sales periods. The warehouse must automatically add clusters when queries are queued and shut them down when demand drops. The company wants to minimize credit consumption while maintaining query performance. Which configuration should the architect implement?

Accepted Answer

Set the warehouse to Auto-scale mode with a scaling policy of Standard.. Auto-scale mode with the Standard scaling policy dynamically adjusts the number of clusters based on query load, adding clusters when queries queue and removing them after idle periods. This provides the necessary performance during peak sales while controlling costs by not over-provisioning. The Standard policy is optimized for performance, making it ideal for time-sensitive workloads that still require cost efficiency.

Answer

Set the warehouse to Auto-scale mode with a scaling policy of Economy.

Answer

Set the warehouse to Maximized mode with a scaling policy of Economy.

Answer

Set the warehouse to Maximized mode with a scaling policy of Standard.

Question 13

An architect is configuring a Snowflake account to use Federated Authentication with Okta. The requirement is that users must authenticate through Okta and receive Snowflake roles based on their Okta group membership. The architect has already configured the SAML integration. What additional step is required to map Okta groups to Snowflake roles?

Accepted Answer

Create a SCIM integration in Snowflake and configure the Okta provisioning connector.. To map Okta groups to Snowflake roles automatically, the architect must set up SCIM integration. SCIM allows Okta to provision users and groups into Snowflake, and Snowflake can then assign roles based on group membership. SAML alone handles authentication but not provisioning. The other options either misunderstand the role of SAML or propose custom solutions that are not standard.

Answer

Configure the Snowflake JDBC driver to pass Okta group information as a session parameter.

Answer

Create a SAML integration with the attribute 'Role' in the Okta SAML assertion.

Answer

Create a stored procedure that queries the Okta API and assigns roles based on group membership.

Question 14

A healthcare organization needs to share patient data with external research partners without copying the data. The partners use their own Snowflake accounts and require read-only access to a specific subset of tables. Which Snowflake feature should the architect use to meet these requirements?

Accepted Answer

Use Snowflake Data Sharing to create a share and grant access to the partners' accounts.. Snowflake Data Sharing enables secure, governed access to data across Snowflake accounts without copying. By creating a share and granting access to the partners' accounts, the healthcare organization can provide read-only access to specific tables. This approach ensures data remains in the provider's account, reducing risk and eliminating data duplication.

Answer

Create a secure view and grant SELECT privileges to the partners' roles.

Answer

Export the data to an external stage and provide the partners with credentials.

Answer

Set up a Snowflake reader account for each partner and grant access to the tables.

Question 15

An architect is designing a Snowflake environment for a financial services company that requires the highest level of security. The company must ensure that data is encrypted at rest and in transit, and that encryption keys are managed by the customer, not Snowflake. Which Snowflake feature should the architect implement?

Accepted Answer

Enable Tri-Secret Secure with a customer-managed key in AWS KMS or Azure Key Vault.. Tri-Secret Secure is the Snowflake feature that allows customers to use their own encryption key, managed in a cloud KMS, in addition to Snowflake's key. This provides customer control over encryption keys and meets the highest security requirements. The other options either do not provide customer-managed keys for Snowflake's internal encryption or are not valid Snowflake features.

Answer

Enable periodic rekeying of Snowflake-managed keys and store the keys in an external HSM.

Answer

Use the ENCRYPTION parameter in the CREATE DATABASE command to specify a customer key.

Answer

Configure Snowflake to use client-side encryption with a customer-provided key before loading data.