Question 39 — AZ-104 Practice Exam 13
A data science team needs a scalable place to land terabytes of mixed video files and raw text that will later be processed by analytics jobs. They don’t want a rigid schema up front, but they do need a flat namespace that apps can read over HTTPS from both on-premises and Azure compute. (Select two answers)
Require a relational database schema before upload to guarantee query performance.
Store the data as objects in Azure Blob Storage to handle large media and text without a fixed schema.
Use Azure Blob Storage to provide HTTPS endpoints that applications can consume from anywhere.
Limit access to SMB shares because object endpoints cannot be called from Azure services.
Correct answers
Store the data as objects in Azure Blob Storage to handle large media and text without a fixed schema.
Use Azure Blob Storage to provide HTTPS endpoints that applications can consume from anywhere.
Explanations
Correct:
Store the data as objects in Azure Blob Storage to handle large media and text without a fixed schema. Object storage is ideal for large media and text ingested before any structured modeling is applied.
Use Azure Blob Storage to provide HTTPS endpoints that applications can consume from anywhere. Blob endpoints support app access over HTTPS for both on-premises and Azure-hosted compute.
Wrong:
Require a relational database schema before upload to guarantee query performance. A fixed schema isn’t required for object storage ingestion; schema-on-read can come later.
Limit access to SMB shares because object endpoints cannot be called from Azure services. Blob object endpoints are routinely consumed by Azure services and on-prem apps.