Explanation
According to the SnowPro Advanced: Architect documents and learning resources, the integration object that should be used to place restrictions on where data may be exported is the security integration. A security integration is a Snowflake object that provides an interface between Snowflake and third-party security services, such as Okta, Duo, or Google Authenticator. A security integration can be used to enforce policies on data export, such as requiring multi-factor authentication (MFA) or restricting the export destination to a specific network or domain. A security integration can also be used to enable single sign-on (SSO) or federated authentication for Snowflake users1.
The other options are incorrect because they are not integration objects that can be used to place restrictions on where data may be exported. Option A is incorrect because a stage integration is not a valid type of integration object in Snowflake. A stage is a Snowflake object that references a location where data files are stored, such as an internal stage, an external stage, or a named stage. A stage is not an integration object that provides an interface between Snowflake and third-party services2. Option C is incorrect because a storage integration is a Snowflake object that provides an interface between Snowflake and external cloud storage, such as Amazon S3, Azure Blob Storage, or Google Cloud Storage. A storage integration can be used to securely access data files from external cloud storage without exposing the credentials, but it cannot be used to place restrictions on where data may be exported3. Option D is incorrect because an API integration is a Snowflake object thatprovides an interface between Snowflake and third-party services that use REST APIs, such as Salesforce, Slack, or Twilio. An API integration can be used to securely call external REST APIs from Snowflake using the CALL_EXTERNAL_API function, but it cannot be used to place restrictions on where data may be exported4. References: CREATE SECURITY INTEGRATION | Snowflake Documentation, CREATE STAGE | Snowflake Documentation, CREATE STORAGE INTEGRATION | Snowflake Documentation, CREATE API INTEGRATION | Snowflake Documentation

Explanation: Option A is the best design to meet the requirements because it uses a materialized view on top of an external table against the S3 bucket in AWS Singapore. A materialized view is a database object that contains the results of a query and can be refreshed periodically to reflect changes in the underlying data1. An external table is a table that references data files stored in a cloud storage service, such as Amazon S32. By using a materialized view on top of an external table, the company can provide access to frequently changing data, keep egress costs to a minimum, and maintain low latency. This is because the materialized view will cache the query results in Snowflake, reducing the need to access the external data files and incur network charges. The materialized view will also improve the query performance by avoiding scanning the external data files every time. The materialized view can be refreshed on a schedule or on demand to capture the changes in the external data files1.
Option B is not the best design because it uses an external table against the S3 bucket in AWS Singapore and copies the data into transient tables. A transient table is a table that is not subject to the Time Travel and Fail-safe features of Snowflake, and is automatically purged after a period of time3. By using an external table and copying the data into transient tables, the company will incur more egress costs and operational overhead than using a materialized view. This is because the external table will access the external data files every time a query is executed, and the copy operation will also transfer data from S3 to Snowflake. The transient tables will also consume more storage space in Snowflake and require manual maintenance to ensure they are up to date.
Option C is not the best design because it copies the data between providers from S3 to Azure Blob storage to collocate, then uses Snowpipe for data ingestion. Snowpipe is a service that automates the loading of data from external sources into Snowflake tables4. By copying the data between providers, the company will incur high egress costs and latency, as well as operational complexity and maintenance of the infrastructure. Snowpipe will also add another layer of processing and storage in Snowflake, which may not be necessary if the external data files are already in a queryable format.
Option D is not the best design because it uses AWS Transfer Family to replicate data between the S3 bucket in AWS Singapore and an Azure Netherlands Blob storage, then uses an external table against the Blob storage. AWS Transfer Family is a service that enables secure and seamless transfer of files over SFTP, FTPS, and FTP to and from Amazon S3 or Amazon EFS5. By using AWS Transfer Family, the company will incur high egress costs and latency, as well as operational complexity and maintenance of the infrastructure. The external table will also access the external data files every time a query is executed, which may affect the query performance.

* A transient table is a type of table in Snowflake that does not have a Fail-safe period and can have a Time Travel retention period of either 0 or 1 day. Transient tables are suitable for temporary or intermediate data that can be easily reproduced or replicated1.
* A temporary table is a type of table in Snowflake that is automatically dropped when the session ends or the current user logs out. Temporary tables do not incur any storage costs, but they are not visible to other users or sessions2.
* A permanent table is a type of table in Snowflake that has a Fail-safe period and a Time Travel retention period of up to 90 days. Permanent tables are suitable for persistent and durable data that needs to be protected from accidental or malicious deletion3.
* In this case, the use case requires loading some data that can be visualized through Tableau. The data is updated every day and the old data is no longer required. Therefore, the best type of table to use in this
* case to optimize cost is a transient table, because it does not incur any Fail-safe costs and it can have a short Time Travel retention period of 0 or 1 day. This way, the data can be loaded and queried by Tableau, and then deleted or overwritten without incurring any unnecessary storage costs.
References: : Transient Tables : Temporary Tables : Understanding & Using Time Travel

* The error message indicates that the schema STAGING does not have time travel data available for the requested timestamp, because the current version of the schema was created on 2021-06-02 23:00:00, which is after the timestamp of 2021-06-01 08:00:00. Therefore, the CLONE statement cannot access
* the historical data of the schema at that point in time.
* Option A is incorrect, because undropping the STAGING schema will not restore the previous version of the schema that was active on 2021-06-01 08:00:00. Instead, it will create a new version of the schema with the same name and no data or objects.
* Option B is incorrect, because modifying the timestamp to 2021-05-01 10:00:00 will not clone the schema as it looked one week ago, but as it looked when it was first created. This may not reflect the desired state of the schema and its objects.
* Option C is correct, because renaming the STAGING schema and performing an UNDROP to retrieve the previous STAGING schema version will restore the schema that was dropped on 2021-06-02
23:00:00. This schema has time travel data available for the requested timestamp of 2021-06-01
08:00:00, and can be cloned using the CLONE statement.
* Option D is incorrect, because cloning can be accomplished by using the UNDROP command to access the previous version of the schema that was active during the proposed time travel period.
References: : Cloning Considerations : Understanding & Using Time Travel : CREATE <object> ... CLONE

A: The Search Optimization Service is designed to accelerate the performance of queries that use filters on large tables. One of the key considerations for its effectiveness is using it with tables where the columns used in the filter conditions have a high number of distinct values, typically in the hundreds of thousands or more.
This is because the service creates a map-reduce-like index on the column to speed up queries that use point lookups or range scans on that column. The more unique values there are, the more effective the index is at narrowing down the search space.References: Snowflake documentation and best practices on the Search Optimization Service, which would be covered under the SnowPro Advanced: Architect certification materials.