Commvault Flex Appliance Deployment Models

For Commvault Flex Appliance, there are effectively three deployment and data path modes that determine how backup data reaches the external S3 storage platform.

1. Storage Accelerator Mode (Direct-to-S3)

Best for: Clients that have direct network connectivity to the S3 storage endpoint.

Characteristics

  • Backup data flows directly from the client to the S3 object store.
  • Flex node manages metadata, indexing, retention, and orchestration.
  • Lowest latency and highest throughput.
  • Reduces load on Flex nodes.
  • Requires network reachability from clients to the S3 endpoint.

2. Proxy Mode (Through Flex Node)

Best for: Segmented networks, security zones, or environments where clients cannot access the S3 endpoint.

Characteristics

  • All backup traffic passes through the Flex node.
  • No direct client access to S3 is required.
  • Simplifies security and network design.
  • Uses Flex node resources for data movement.
  • Common in highly secured enterprise environments.

3. Hybrid Mode

Best for: Large enterprises with multiple sites, network zones, or mixed workload requirements.

Characteristics

  • Some workloads use direct-to-S3 acceleration.
  • Other workloads use the Flex node as a proxy.
  • Allows optimization based on network topology and security requirements.
  • Often the preferred design in large, distributed environments.

Flex Deployment Architecture Options

Separately from the data path modes, Flex Appliance supports the following deployment architectures.

Architecture Description
2-Node Flex Cluster Minimum supported deployment; provides resiliency and availability.
Scale-Out Flex Cluster Additional Flex nodes are added to increase throughput and capacity management.
Independent Compute/Storage Scaling Flex nodes and external S3 storage can be expanded independently.

Flex Appliance is always deployed with:

  • Two or more Flex nodes.
  • Certified Dell, HPE, or Lenovo servers.
  • Certified external S3 storage.
  • 100Gb networking recommended for large-scale deployments.
×

Loading...