Universal Containers (UC) wants to test code against a subset of production data that is under 5 GB. Additionally, UC wants to refresh this sandbox…

A Partial Copy sandbox is the ideal choice for this scenario. It is specifically designed to contain the organization's metadata plus a defined subset of production data, which is selected using a sandbox template. This type of sandbox has a 5 GB data storage limit, aligning with the requirement. Its 5-day refresh interval allows for a weekly refresh cycle, such as refreshing every weekend, making it the only option that satisfies all of Universal Containers' requirements for data content, size, and refresh frequency.

A. Developer Pro: This sandbox type only copies metadata from the production environment, not a subset of its data records.

C. Full Copy: This sandbox contains a complete copy of all production data, not a subset, and its 29-day refresh interval prevents weekly refreshes.

D. Developer: Like the Developer Pro sandbox, this type only copies metadata and does not include any production data records by default.

1. Salesforce Help: Sandbox Types and Templates. This official documentation provides a comparison table detailing the features of each sandbox type.

Partial Copy Sandbox: The table specifies its content as "All of your organization's metadata and a sample of your production org's data that you define by a sandbox template," its storage limit as "5 GB of data," and its refresh interval as "5 days."

Full Sandbox: The table notes its refresh interval is "29 days."

Developer and Developer Pro Sandboxes: The table confirms their content is limited to "a copy of your production org’s configuration (metadata)."

2. Salesforce Developer Documentation: Sandboxes: Staging Environments for Customizing and Testing. This guide reinforces the use cases for each sandbox type.

Section: Sandbox Types: It states, "A Partial Copy sandbox is a Developer sandbox plus the data that you define in a sandbox template... Use a Partial Copy sandbox for quality assurance tasks such as user acceptance testing, integration testing, and training." This aligns with the need to test code against a representative data set.