11.2.2: File Storage

File storage is the traditional hierarchical storage model where data is organized into files placed in directories (folders) and subdirectories. Users or applications access data via a file path (e.g., /project/docs/report.pdf), and the system uses a file system (such as NTFS, ext4, etc.) to map these paths to physical blocks on disk. File storage systems maintain metadata like filenames, sizes, timestamps, and permissions in a central directory structure. In a cloud context, cloud file storage often refers to managed Network Attached Storage (NAS) or file server instances provided as a service, enabling multiple cloud compute instances to share files over standard protocols (like NFS or SMB). Examples include Amazon EFS (Elastic File System), Azure Files, or Google Cloud Filestore, which present a POSIX-compliant file system interface accessible over the network.

The main advantage of file storage is its familiarity and ease of use for both users and applications. Many legacy applications expect a file system interface and cannot easily be rewritten for object or block storage. File storage allows concurrent access and file locking, enabling collaborative use (multiple users editing files in shared directories) and compatibility with existing software. Cloud file storage services typically manage the underlying infrastructure (scaling capacity, maintaining redundancy, etc.), so that users get a simple shared drive that grows on demand. For instance, Azure Files provides a fully managed SMB file share that can be mounted by VMs or accessed via APIs, making it easy to migrate on-premises file server workloads to the cloud.

However, file storage can be less scalable than object storage. Because it uses a hierarchical namespace and often a single metadata server (to coordinate directory structures and locks), extremely large-scale deployments might hit performance bottlenecks on metadata operations (listing directories, opening thousands of files, etc.). Cloud providers mitigate this by behind-the-scenes sharding of storage or offering guidance on the maximum throughput. File storage is typically best for moderate-scale, shared data scenarios – for example, a content management system, home directories for users, or small team file shares. It also sees use in lift-and-shift cloud migrations, where an application that expects a local file system can be rehosted on a VM with attached cloud file storage without significant code changes.

Use cases: Common uses for cloud file storage include enterprise shared drives, content repositories, and application file storage for workloads that need filesystem semantics (like file locking or POSIX compliance). For instance, a web application might store user-uploaded files on an NFS share (AWS EFS) so that all application servers have access to the same files. In data analytics, some frameworks require distributed file access – a cloud file system can serve as a convenient solution for intermediate data that needs to be accessed by multiple processing nodes. File storage is also often used for legacy application support in the cloud, where rewriting to use object storage is not feasible; the application can instead use a mounted file path while the cloud service handles durability behind the scenes.

---