Amazon Web Services (AWS) today announced the general availability of AWS Interconnect, a new suite of managed private connectivity services designed to streamline the complex task of connecting disparate cloud environments and on-premises infrastructure. The launch introduces two key capabilities: AWS Interconnect—multicloud, enabling direct, private connections between Amazon Virtual Private Clouds (Amazon VPCs) and VPCs on other major cloud providers, and AWS Interconnect—last mile, simplifying the establishment of high-speed, private links from enterprise locations to AWS.
This move by AWS directly addresses a growing trend in enterprise IT: the adoption of multicloud strategies. Organizations are increasingly distributing their workloads across various cloud platforms to leverage specialized services, meet stringent data residency requirements, or accommodate teams with existing preferences for different cloud ecosystems. However, this distributed approach has historically presented significant networking challenges, requiring substantial manual effort in managing VPN tunnels, coordinating with colocation facilities, and configuring complex third-party network fabrics. The operational overhead associated with these tasks diverts valuable IT resources from core business initiatives.
AWS Interconnect aims to alleviate these pain points by offering a fully managed, turnkey solution. The service provides dedicated bandwidth for private, high-speed network connections to and from AWS, spanning both hybrid and multicloud architectures. Users can configure resilient, end-to-end connectivity through the AWS Console with a few clicks, selecting their desired location, partner or cloud provider, AWS Region, and bandwidth requirements. This streamlined process eliminates the need for manual network configuration and the complexities of partner discovery.
Unveiling AWS Interconnect—Multicloud
AWS Interconnect—multicloud establishes a private, managed Layer 3 connection between an organization’s AWS environment and those of other cloud providers. Initial support is available for Google Cloud, with Microsoft Azure and Oracle Cloud Infrastructure (OCI) slated for integration later in 2026. Traffic transmitted via this service flows exclusively over the AWS global backbone and the partner cloud provider’s private network, ensuring it never traverses the public internet. This architecture guarantees predictable latency, consistent throughput, and isolation from internet congestion, all without the burden of managing physical infrastructure.
Security is a fundamental aspect of AWS Interconnect—multicloud, with IEEE 802.1AE MACsec encryption implemented by default on the physical links between AWS and partner cloud provider routers at interconnection facilities. This encryption is automatic and does not require separate configuration by the user. Organizations should, however, consult the encryption documentation of their respective cloud providers to ensure compliance with specific security and regulatory requirements. Resiliency is also a core design principle, with each connection spanning multiple logical links distributed across at least two physical facilities to prevent service disruptions in the event of a single device or building failure.
For monitoring, AWS Interconnect—multicloud integrates seamlessly with Amazon CloudWatch. Each connection includes a Network Synthetic Monitor that provides insights into round-trip latency, packet loss, and bandwidth utilization metrics, crucial for capacity planning and performance optimization.
AWS has also adopted an open approach to fostering collaboration for its multicloud offering. The underlying specification for AWS Interconnect—multicloud has been published on GitHub under the Apache 2.0 license, inviting other cloud service providers to collaborate. To become an AWS Interconnect partner, providers must meet AWS’s stringent operational requirements, including resiliency standards, support commitments, and service level agreements, in addition to implementing the technical specification.
Simplified Provisioning: A Step-by-Step Example
The provisioning process for AWS Interconnect—multicloud is designed for speed and simplicity, taking mere minutes. A user can initiate the connection request directly from the AWS Direct Connect console, selecting the AWS Interconnect section.
The process typically involves the following steps:
- Request an Interconnect: Within the AWS Management Console, navigate to AWS Direct Connect and then AWS Interconnect. Select "Create" and choose the desired cloud provider, such as Google Cloud. Specify the source and destination AWS Regions and the corresponding cloud provider regions.
- Specify Bandwidth and Project Details: Enter a description for the connection, select the required bandwidth, choose the Direct Connect Gateway to attach, and provide the cloud provider’s project ID.
- Receive Activation Key: Upon confirming the request, the AWS console generates an activation key. This key is essential for validating the connection request on the cloud provider’s side.
- Configure on Cloud Provider Platform: Using the activation key, the user proceeds to the chosen cloud provider’s platform. For instance, when connecting to Google Cloud Platform (GCP), this might involve using the
gcloudcommand-line interface to create a transport and then establish a VPC peering connection. This step ensures that the network routes are propagated correctly between the two environments. - Associate Gateway and Route Table: On the AWS side, the user associates the new gateway with their Virtual Private Gateway (VGW) or Transit Gateway. A final step involves adding a route entry in the VPC’s route tables to direct traffic destined for the cloud provider’s IP address range through the Virtual Gateway.
A practical demonstration involved connecting a single AWS VPC to a Google Cloud VPC using a Direct Connect Gateway. The process, initiated from the AWS console, involved selecting Google Cloud as the provider, specifying the eu-central-1 AWS Region and europe-west3 Google Cloud Region, and entering the Google Cloud project ID. An activation key was then generated.
On the GCP side, the gcloud network-connectivity transports create command was used with the activation key, specifying the network and advertised routes. Subsequently, a gcloud compute networks peerings create command established the peering between the GCP VPC and the newly created transport. Verification in both the GCP console and the AWS Interconnect console confirmed the connection’s availability.
The final configuration steps involved associating the Direct Connect gateway in AWS and updating the VPC route tables to direct traffic towards the GCP IP address range via the Virtual Gateway. Testing the connection involved launching compute instances in both AWS and GCP, starting a simple web server on AWS, and then using curl from the GCP instance to access the AWS web server via its private IP address, confirming successful private connectivity.
AWS Interconnect—Last Mile: Bridging On-Premises to the Cloud
Complementing the multicloud offering, AWS Interconnect—last mile extends similar managed connectivity benefits to on-premises data centers and remote branch offices. This capability allows organizations to establish high-speed, private connections to AWS through their existing network providers’ last-mile infrastructure, directly managed via the AWS Management Console.
The onboarding process mirrors that of the multicloud service: users select a participating network provider, authenticate, and specify connection endpoints and desired bandwidth. An activation key is generated by AWS, which is then used on the provider’s console to finalize the configuration. AWS Interconnect—last mile automatically provisions four redundant connections across two physical locations, configures BGP routing, and enables MACsec encryption and Jumbo Frames by default. This ensures a resilient private connection to AWS that adheres to best practices without requiring manual configuration of networking components.
AWS Interconnect—last mile supports bandwidths ranging from 1 Gbps to 100 Gbps, with the flexibility to adjust bandwidth from the console without re-provisioning. The service includes a 99.99% availability Service Level Agreement (SLA) up to the Direct Connect port and integrates CloudWatch Network Synthetic Monitor for connection health monitoring. Similar to its multicloud counterpart, AWS Interconnect—last mile attaches to a Direct Connect Gateway, which can then connect to VGW, Transit Gateway, or AWS Cloud WAN deployments.
Lumen Technologies is highlighted as an initial partner for AWS Interconnect—last mile, with AT&T and Megaport expected to join as partners soon. This collaboration aims to simplify the often complex last-mile connectivity challenges that can impede cloud adoption.
Pricing and Availability
The pricing for both AWS Interconnect—multicloud and AWS Interconnect—last mile is based on a flat hourly rate for the provisioned capacity, billed prorata by the hour. Customers select the bandwidth tier that best suits their workload requirements.
Pricing for AWS Interconnect—multicloud varies depending on the region pair. For example, a connection between US East (N. Virginia) and Google Cloud N. Virginia will have a different cost than a connection between US East (N. Virginia) and a more geographically distant region. When using AWS Cloud WAN, the global routing model can influence total deployment costs due to potential multi-region traffic traversal. Detailed pricing information can be found on the respective AWS Interconnect pricing pages for multicloud and last mile.
As of the announcement, AWS Interconnect—multicloud is available in five region pairs:
- US East (N. Virginia) to Google Cloud N. Virginia
- US West (N. California) to Google Cloud Los Angeles
- US West (Oregon) to Google Cloud Oregon
- Europe (London) to Google Cloud London
- Europe (Frankfurt) to Google Cloud Frankfurt
Microsoft Azure support is anticipated later in 2026, with Oracle Cloud Infrastructure (OCI) also listed as a future integration.
AWS Interconnect—last mile is launching initially in US East (N. Virginia) with Lumen Technologies as the first partner. Additional partners and regions are planned for future expansion.
Organizations interested in leveraging AWS Interconnect can access the service through the AWS Direct Connect console and selecting AWS Interconnect from the navigation menu.
Broader Impact and Future Architectures
The introduction of AWS Interconnect signifies a strategic move by AWS to simplify multicloud networking, a growing imperative for large enterprises. By abstracting away the underlying network complexities, AWS empowers organizations to focus on innovation and application development rather than infrastructure management.
Reference Architectures for Scalability: For deployments involving multiple VPCs within a single AWS region, AWS Transit Gateway can serve as a centralized routing hub, connecting them through a single Interconnect attachment. This facilitates traffic segmentation, consistent routing policy enforcement, and integration with AWS Network Firewall for traffic inspection.
At a global scale, for workloads distributed across multiple AWS Regions and other cloud environments, AWS Cloud WAN offers a unified model. This allows any region within an organization’s network to connect to any Interconnect attachment globally, supported by centralized policy management and segment-based routing. Detailed guidance on building resilient and scalable multicloud connectivity architectures is available in a companion blog post by AWS colleagues Alexandra and Santiago.
The proactive approach to publishing specifications on GitHub and the phased rollout of provider integrations suggest a commitment to fostering a more interconnected and interoperable cloud ecosystem. This initiative is likely to be a key enabler for enterprises looking to maximize the benefits of their multicloud investments while maintaining robust security, reliability, and performance.
