Skip to content

Choosing the Right Connector🔗

The correct DSX-Connect connector is determined by how your files are hosted and accessed.

Do not start by asking “Which connector should I use?”

Start by asking:

How are your files stored today?

Step 1: Identify the Storage Model🔗

Most enterprise file repositories fall into one of three categories.

1️⃣ Object Storage (API-Based)🔗

Examples:

  • Amazon S3
  • Azure Blob Storage
  • Google Cloud Storage

Characteristics:

  • Accessed via REST API
  • Cloud-native
  • Horizontally scalable
  • No mounted storage required

Use the corresponding object storage connector.

These connectors communicate directly with the storage provider API.

2️⃣ Network-Served Filesystems🔗

Examples:

  • NFS
  • SMB / CIFS
  • NAS appliances (NetApp, Isilon)
  • Azure Files
  • AWS EFS
  • CephFS
  • Enterprise file shares

These systems:

  • Serve files over the network
  • Support multiple clients
  • Act as centralized storage services

Even though they are “filesystems,” they behave like storage services.

Use the Filesystem connector.

Deployment pattern:

  • Docker → Mount on host, then bind-mount into container
  • Kubernetes → Mount via PersistentVolumeClaim (RWX recommended)

This is the most common enterprise filesystem use case.

3️⃣ Node-Local Storage (Edge / Single Server)🔗

Examples:

  • /mnt/data on a server
  • Local desktop folders
  • Edge devices

Characteristics:

  • Not centrally served
  • Bound to a specific machine
  • Not inherently horizontally scalable

Use the Filesystem connector.

However:

  • Docker Compose is often simpler than Kubernetes for node-local storage.
  • In Kubernetes, hostPath ties the connector to a node.

This pattern is best suited for:

  • Edge deployments
  • Single-node clusters
  • Development environments

Architectural Summary🔗

Storage Type Recommended Connector Notes
Object Storage S3 / Azure / GCS connector Cloud-native, API-driven
Networked Filesystem Filesystem connector Mount via PVC or bind mount
Node-Local Disk Filesystem connector Edge / dev only

Key Insight🔗

The Filesystem connector is not a special-case connector.

The storage backend determines:

  • Portability
  • Scheduling behavior (Kubernetes)
  • Scalability
  • Deployment complexity

In production Kubernetes environments, network-backed storage is recommended.

When in Doubt🔗

If your storage is accessed by multiple systems over the network, it is almost always a served storage system, and the Filesystem connector should be deployed using PVC-backed storage.