As we speak, we’re asserting the final availability of AWS Interconnect – multicloud, a managed personal connectivity service that connects your Amazon Digital Personal Cloud (Amazon VPC) on to VPCs on different cloud suppliers. We’re additionally introducing AWS Interconnect – final mile, a brand new functionality that simplifies how you identify high-speed, personal connections to AWS out of your department places of work, knowledge facilities, and distant areas by your current community suppliers.
Giant enterprises more and more run workloads throughout a number of cloud suppliers, whether or not to make use of specialised providers, meet knowledge residency necessities, or help groups which have standardized on totally different suppliers. Connecting these environments reliably and securely has traditionally required vital coordination: managing VPN tunnels, working with colocation amenities, and configuring third-party community materials. The result’s that your networking workforce spends time on undifferentiated heavy lifting as an alternative of specializing in the purposes that matter to your small business.
AWS Interconnect is the reply to those challenges. It’s a managed connectivity service that simplifies connectivity into AWS. Interconnect supplies you the flexibility to determine personal, high-speed community connections with devoted bandwidth to and from AWS throughout hybrid and multicloud environments. You’ll be able to configure resilient, end-to-end connectivity with ease in a number of clicks by the AWS Console by deciding on your location, associate, or cloud supplier, most well-liked Area, and bandwidth necessities, eradicating the friction of discovering companions and the complexity of guide community configurations.
It comes with two capabilities: multicloud connectivity between AWS and different cloud suppliers, and last-mile connectivity between AWS and your personal on-premises networks. Each capabilities are constructed on the identical precept: a totally managed, turnkey expertise that removes the infrastructure complexity out of your workforce.
AWS Interconnect – multicloud
AWS Interconnect – multicloud offers you a personal, managed Layer 3 connection between your AWS surroundings and different cloud suppliers, beginning with Google Cloud, and Microsoft Azure coming later in 2026. Site visitors flows solely over the AWS world spine and the associate cloud’s personal community, so it by no means traverses the general public web. This implies you get predictable latency, constant throughput, and isolation from web congestion with out having to handle any bodily infrastructure your self.
Safety is inbuilt by default. Each connection makes use of IEEE 802.1AE MACsec encryption on the bodily hyperlinks between AWS routers and the associate cloud supplier’s routers on the interconnection amenities. You don’t must configure these individually. Notice that every cloud supplier manages encryption independently by itself spine, so you need to evaluate the encryption documentation to your particular deployment to confirm it meets your compliance necessities. Resiliency can be inbuilt: every connection spans a number of logical hyperlinks distributed throughout a minimum of two bodily amenities, so a single system or constructing failure doesn’t interrupt your connectivity.
For monitoring, AWS Interconnect – multicloud integrates with Amazon CloudWatch. You get a Community Artificial Monitor included with every connection to trace round-trip latency and packet loss, and bandwidth utilization metrics to help capability planning.
AWS has printed the underlying specification on GitHub underneath the Apache 2.0 license, offering any cloud service supplier the chance to collaborate with AWS Interconnect – multicloud. To turn out to be an AWS Interconnect associate, cloud suppliers should implement the technical specification and meet AWS operational necessities, together with resiliency requirements, help commitments, and repair stage agreements.
The way it works
Provisioning a connection takes minutes. I create the connection from the AWS Direct Join console. I begin from the AWS Interconnect part and choose Google Cloud because the supplier. I choose my supply and vacation spot areas. I specify bandwidth, and supply my Google Cloud challenge ID. AWS generates an activation key that I exploit on the Google Cloud aspect to finish the connection. Routes propagate mechanically in each instructions, and my workloads can begin exchanging knowledge shortly after.
For this demo, I begin with a single VPC and I join it to a Google Cloud VPC. I exploit a Direct Join Gateway. It’s the best path: one connection, one attachment, and my workloads on either side can begin speaking to one another in minutes.
Step 1: request an interconnect within the AWS Administration Console.
I navigate to AWS Direct Join, AWS Interconnect and I choose Create. I first select the cloud supplier I need to hook up with. On this instance, Google Cloud.
Then, I select the AWS Area (eu-central-1) and the Google Cloud Area (europe-west3).
On step 3, I enter Description,I select the Bandwidth, the Direct Join gateway to connect, and the ID of my Google Cloud challenge.

After reviewing and confirming the request, the console offers me an activation key. I’ll use that key to validate the request on the Google cloud aspect.

Step 2: create the transport and VPC Peering sources on my Google Cloud Platform (GCP) account.
Now that I’ve the activation key, I proceed the method on the GCP aspect. On the time of this writing, no web-based console was accessible. I select to make use of the GCP command line (CLI) as an alternative. I be aware of the CIDR vary within the GCP VPC subnet within the europe-west3 area. Then, I open a Terminal and kind:
gcloud network-connectivity transports create aws-news-blog
--region=europe-west3
--activation-key=${ACTIVATION_KEY}
--network=default
--advertised-routes=10.156.0.0/20
Create request issued for: [aws-news-blog]
...
peeringNetwork: tasks/oxxxp-tp/world/networks/transport-9xxxf-vpc
...
state: PENDING_CONFIG
updateTime: '2026-03-19T09:30:51.103979219Z'
It takes a few minutes for the command to finish. As soon as the command returns, I create a peering between my GCP VPC and the brand new transport I simply created. I can do this within the GCP console or with the gcloud command line. As a result of I used to be utilizing the Terminal for the earlier command, I continued with the command line:
gcloud compute networks peerings create aws-news-blog
--network=default
--peer-network=tasks/oxxxp-tp/world/networks/transport-9xxxf-vpc
--import-custom-routes
--export-custom-routes
The community identify is the identify of my GCP VPC. The peer community is given within the output of the earlier command.
As soon as accomplished, I can confirm the peering within the GCP console.

Within the AWS Interconnect console, I confirm the standing is accessible.
Within the AWS Direct Join console, underneath Direct Join gateways, I see the attachment to the brand new interconnect.

Step 3: affiliate the brand new gateway on the AWS aspect
I choose Gateway associations and Affiliate gateway to connect the Digital Personal Gateway (VGW) that I created earlier than beginning this demo (take note of use a VGW in the identical AWS Area because the interconnect)

You don’t must configure the community routing on the GCP aspect. On AWS, there’s a remaining step: add a route entry in your VPC Route tables to ship all site visitors to the GCP IP deal with vary by the Digital Gateway.

As soon as the community setup is completed. I begin two compute cases, one on AWS and one on GCP.
On AWS, I confirm the Safety Group accepts ingress site visitors on TCP:8080. I hook up with the machine and I begin a minimal internet server:
python3 -c
"from http.server import HTTPServer, BaseHTTPRequestHandler
class H(BaseHTTPRequestHandler):
def do_GET(self):
self.send_response(200);self.end_headers()
self.wfile.write(b'Hi there AWS World!nn')
HTTPServer(('',8080),H).serve_forever()"On the GCP aspect, I open a SSH session to the machine and I name the AWS internet server by its personal IP deal with.

Et voilà! I’ve a personal community route between my two networks, solely managed by the 2 Cloud Service Suppliers.
Issues to know
There are a few configuration choices that you need to take into accout:
- When connecting networks, take note of the IP addresses vary on either side. The GCP and AWS VPC ranges can’t overlap. For this demo, the default vary on AWS was
172.31.0.0/16and the default on GCP was10.156.0.0/20. I used to be capable of proceed with these default values. - You’ll be able to configure IPV4, IPV6, or each on all sides. You should choose the identical possibility on either side.
- The Most Transmission Unit (MTU) have to be the identical on each VPC. The default values for AWS VPCs and GCP VPCs aren’t. MTU is the biggest packet dimension, in bytes, {that a} community interface can transmit with out fragmentation. Mismatched MTU sizes between peered VPCs trigger packet drops or fragmentation, resulting in silent knowledge loss, degraded throughput, and damaged connections throughout the interconnect.
- For extra particulars, discuss with the GCP Associate Cross Cloud Interconnect and the AWS Interconnect Consumer Information.
Reference architectures
When your deployment grows and you’ve got a number of VPCs in a single area, AWS Transit Gateway offers you a centralized routing hub to attach all of them by a single Interconnect attachment. You’ll be able to section site visitors between environments, apply constant routing insurance policies, and combine AWS Community Firewall if you could examine what crosses the cloud boundary.
And if you’re working at world scale, with workloads unfold throughout a number of AWS Areas and a number of Google Cloud environments, AWS Cloud WAN extends that very same mannequin internationally. Any area in your community can attain any Interconnect attachment globally, with centralized coverage administration and segment-based routing that applies constantly all over the place you use.
My colleagues Alexandra and Santiago documented these reference architectures of their weblog submit: Construct resilient and scalable multicloud connectivity architectures with AWS Interconnect – multicloud.
AWS Interconnect – final mile
Based mostly on the identical structure and design as AWS Interconnect – multicloud, AWS Interconnect – final mile supplies the flexibility to attach your on-premises or distant location to AWS by a taking part community supplier’s last-mile infrastructure, immediately from the AWS Administration Console.
The onboarding course of mirrors AWS Interconnect – multicloud: you choose a supplier, authenticate, and specify your connection endpoints and bandwidth. AWS generates an activation key that you just present within the supplier console to finish the configuration. AWS Interconnect – final mile mechanically provisions 4 redundant connections throughout two bodily areas, configures BGP routing, and prompts MACsec encryption and Jumbo Frames by default. The result’s a resilient personal connection to AWS that aligns with greatest practices, with out requiring you to manually configure networking parts.

AWS Interconnect – final mile helps bandwidths from 1 Gbps to 100 Gbps, and you may regulate bandwidth from the console with out reprovisioning. The service features a 99.99% availability SLA as much as the Direct Join port and bundles CloudWatch Community Artificial Monitor for connection well being monitoring. Identical to AWS Interconnect – multicloud, AWS Interconnect – final mile attaches to a Direct Join Gateway, which connects to your Digital Personal Gateway, Transit Gateway, or AWS Cloud WAN deployment. For extra particulars, discuss with the AWS Interconnect Consumer Information.
Scott Yow, SVP Product at Lumen Applied sciences, wrote:
By combining AWS Interconnect – final mile with Lumen fiber community and Cloud Interconnect, we simplify the last-mile complexity that always slows cloud adoption and allow a quicker, and extra resilient path to AWS for purchasers.
Pricing and availability
AWS Interconnect – multicloud and AWS Interconnect – final mile pricing is predicated on a flat hourly charge for the capability you request, billed prorata by the hour. You choose the bandwidth tier that matches your workload wants.
AWS Interconnect – multicloud pricing varies by area pair: a connection between US East (N. Virginia) and Google Cloud N. Virginia is priced in a different way from a connection between US East (N. Virginia) and a extra distant area. Once you use AWS Cloud WAN, the worldwide any-to-any routing mannequin means site visitors can traverse a number of areas, which impacts the overall value of your deployment. I like to recommend reviewing the AWS Interconnect – multicloud pricing web page and AWS Interconnect – final mile pricing web page for the complete charge card by area pair and capability tier earlier than sizing your connection.
AWS Interconnect – multicloud is out there right now in 5 area 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, and Europe (Frankfurt) to Google Cloud Frankfurt. Microsoft Azure help is coming later in 2026.
AWS Interconnect – final mile is launching in US East (N. Virginia) with Lumen because the preliminary associate. Further companions, together with AT&T and Megaport, are in progress, and extra areas are deliberate.
To get began with AWS Interconnect, go to the AWS Direct Join console and choose AWS Interconnect from the navigation menu.
I’d love to listen to the way you’re utilizing AWS Interconnect in your surroundings. Go away a remark beneath or attain out by the AWS re:Publish neighborhood.
— seb
Up to date on April 15 – We mounted unsuitable hyperlink for pricing pages.



