You need to recommend a non-relational data store that is optimized for storing and retrieving text files, videos, audio streams, and virtual disk images. The data store must store data, some metadata

Columnar databases are optimized for analytical workloads, storing data in columns to efficiently query aggregates over large datasets, which is not ideal for storing and retrieving diverse unstructured files.

Key/value stores are optimized for storing simple key-value pairs and fast lookups, but are not primarily designed for handling large, unstructured files with associated metadata efficiently at the object level.

Document databases store semi-structured data as documents (e.g., JSON), which are suitable for flexible schema data but not typically optimized for storing large binary files like videos or virtual disk images themselves, often storing references to such files instead.

Object storage, such as Azure Blob Storage, is optimized for storing large, unstructured data objects like text files, videos, audio streams, and virtual disk images, and it inherently stores data alongside customizable metadata and provides a unique identifier for each object.