Skip to main contentCloud Links provide private, high-performance connectivity between your physical ports and a public cloud provider. Use them to:
- Establish private, low-latency paths between your network and a public cloud (AWS, Azure, GCP).
- Build a resilient, scalable foundation for hybrid and multi-cloud architectures.
- Bypass the internet to provide enhanced security, reliability, and performance.
Cloud Links come through your provisioned ports and leverage the Zayo DynamicLink fabric for reliability, performance, and security.
Core concepts
- A-side / Z-side: A connection links two sides. The A-side is your initiating port; the Z-side is the destination.
- Bandwidth: You allocate bandwidth for the connection. Ensure both ends have sufficient available capacity.
- VLANs: VLANs segment traffic on a port. Each connection uses selected VLANs to isolate and identify the connection.
- Status lifecycle: Ordering, Pending/Provisioning and Available/Active. Status is visible in the portal and (for clouds) within the provider console.
- BGP: For cloud connections, BGP enables dynamic route exchange once the NNI is established.
Design guidance
- Redundancy: Build dual links on separate ports/locations for high availability
- Capacity planning: Size bandwidth to peak traffic; monitor and adjust as needs evolve
- Segmentation: Use distinct VLANs per environment or application to simplify operations
- Observability: Enable monitoring to track utilization, latency, and loss end-to-end
Connection types
Port-to-cloud connectivity
Connect directly to public cloud providers via Network-to-Network Interface (NNI):
- AWS Direct Connect: Create the connection in DynamicLink portal, then approve it in the AWS console.
- GCP Cloud Interconnect: Start in the Google Cloud Console to create the interconnect and obtain a pairing key. You will use this key to create the connection in DynamicLink.
- Azure ExpressRoute: Create the circuit in Azure to obtain a service key. You will use this key to create the connection in DynamicLink.
Bandwidth options
AWS, Azure, and Google Cloud allow the following bandwidths:
- 50 Mbps: Suitable for small workloads and testing
- 100 Mbps: Good for development environments
- 200 Mbps: Appropriate for small production workloads
- 300 Mbps: Medium-sized applications
- 400 Mbps: Larger applications with moderate traffic
- 500 Mbps: High-traffic applications
- 1 Gbps: Enterprise applications
- 2 Gbps: Large enterprise workloads
- 5 Gbps: High-bandwidth applications
- 10 Gbps: Maximum hosted connection bandwidth
Cloud-to-cloud connectivity
You can create a Layer 2 virtual circuit between two cloud locations or two different cloud providers. This is useful if you need Layer 2 connectivity rather than Layer 3 connectivity (such as with Cloud Routers).
Multi-cloud network fabrics allow AWS, Azure, and GCP resources to communicate as if on the same Layer-2 domain. This unified approach enables true workload portability across providers, removing the need for repeated reconfiguration.