Cross-Cloud Interconnect overview

Cross-Cloud Interconnect is a product that helps you establish high-bandwidth dedicated connectivity between Google Cloud and another cloud service provider.

When you buy Cross-Cloud Interconnect, Google provisions a dedicated physical connection between the Google network and that of another cloud service provider. You can use this connection to peer your Google Virtual Private Cloud (VPC) network with your network that's hosted by a supported cloud service provider.

Cross-Cloud Interconnect configuration.
Cross-Cloud Interconnect configuration (click to enlarge).

Google supports the connection up to the point where it reaches the network of your other cloud service provider. Google does not guarantee uptime from the other cloud service provider and cannot create a support ticket on your behalf. However, with your permission Google Cloud Support can communicate directly with your other cloud provider's support team to expedite issue resolution.

Cross-Cloud Interconnect helps you avoid some of the common pain points associated with multicloud configuration, as described in the following section.

Benefits

This section describes the benefits of multicloud in general and of Cross-Cloud Interconnect in particular.

Integrated multicloud strategy

Cross-Cloud Interconnect supports your adoption of an integrated multicloud strategy. Adopting a multicloud architecture lets you do the following:

  • Avoid being locked in with a single vendor.

  • Store data in one cloud while hosting business logic in another.

  • Avoid downtime if one cloud has an outage.

  • Use a second cloud for disaster recovery.

  • Maximize business insights by analyzing data in multiple clouds.

Reduced complexity

Without Cross-Cloud Interconnect, the options for setting up connectivity are limited, and all are relatively complex.

One option is to deploy your own router in a colocation facility and then connect it to the networks of your cloud service providers. In general, this approach can be expensive and time consuming.

Another option is to contract with a third party to establish connectivity. However, it can be a hassle to select a vendor, or potentially multiple vendors, and then negotiate one or more contracts. When you use this approach, you also have to invest time learning about the systems that are specific to your vendors.

When you use Cross-Cloud Interconnect, you don't have to deploy your own hardware, and you eliminate the need to work with third parties.

Site-to-site data transfer

You can use Cross-Cloud Interconnect as part of a site-to-site data transfer strategy. Site-to-site data transfer is a feature of Network Connectivity Center that lets you use the Google network as a wide area network (WAN).

With this feature, you connect your external networks to Google Cloud by using supported connectivity resources. You then associate each connectivity resource with a Network Connectivity Center spoke and enable the spokes for site-to-site data transfer. At that point, you can conduct data transfer between the sites.

You can use this feature to do the following:

  • Connect your cloud networks: Suppose you have a network hosted by Microsoft Azure and another hosted by Amazon Web Services (AWS). In this scenario, you can establish one pair of Cross-Cloud Interconnect connections to reach the Azure network and another to reach the AWS network. After configuring Network Connectivity Center spokes, you can use the Google network to transfer data between your Azure and AWS networks.

  • Connect an on-premises network to other clouds: Suppose you have the setup that's described in the preceding bullet point, but you also have offices in New York and Sydney. In this scenario, you can establish connectivity to your offices by using resources such as Dedicated Interconnect VLAN attachments or Cloud VPN (HA VPN) tunnels. After creating spokes, you can use the Google network to transfer data between either of your offices and your Azure and AWS networks.

Site-to-site data transfer is supported only in certain locations.

Encryption

Cross-Cloud Interconnect supports MACsec for Cloud Interconnect encryption for additional security. You can use MACsec for Cloud Interconnect to help secure traffic in your Cross-Cloud Interconnect connections. For more information, see MACsec for Cloud Interconnect overview.

Supported cloud service providers

Google supports the following cloud service providers for use with Cross-Cloud Interconnect:

  • Amazon Web Services (AWS)

  • Microsoft Azure

  • Oracle Cloud Infrastructure (OCI)

  • Alibaba Cloud

Cross-Cloud Interconnect and this documentation refer to such cloud service providers as your remote cloud service provider or remote cloud.

Comparison to Partner Cross-Cloud Interconnect for OCI

While Cross-Cloud Interconnect lets you turn up dedicated connectivity between Google Cloud and OCI, Partner Cross-Cloud Interconnect for OCI lets you connect any Google Cloud and OCI resources privately in available paired locations. Use the following table to choose the option that works best for your needs:

Feature Cross-Cloud Interconnect Partner Cross-Cloud Interconnect for OCI
Tenancy

Connections are fully owned by you and can't be shared across organizations.

Google Cloud and OCI own ports; connections can be used by multiple customers.

Supported speeds

10 Gbps, 100 Gbps.

1 Gbps, 2 Gbps, 5 Gbps, 10 Gbps, 20 Gbps, 50 Gbps.

Cost

You pay for ports and attachments in both clouds and data transfer out for traffic leaving Google Cloud.

You pay for Google Cloud partner attachments and OCI virtual circuits. There is no data transfer fee.

Performance

Optimal latency.

Optimal latency.

Support

Google supports the connection all the way to the OCI demarcation.

Google Cloud and OCI partner on supporting the solution end to end.

Google support for Cross-Cloud Interconnect

The following diagram shows the point of physical cabling that Google support is responsible for. For any other issues, contact the other cloud service provider that you are using.

Cross-Cloud Interconnect physical cabling support.
Cross-Cloud Interconnect physical cabling support (click to enlarge).

Capacity

Cross-Cloud Interconnect connections are available in two sizes: 10 Gbps or 100 Gbps.

Setup process

To start the setup process, you identify supported locations where you want Google to place your connections. Then you purchase primary and redundant Cross-Cloud Interconnect ports. You also buy primary and redundant ports from your remote cloud service provider.

Following your purchase from your remote cloud service provider, they provide you with documentation that authorizes you to connect to their router. You send this documentation to Google. At that point, Google can provision your connections. For each connection, Google physically connects one of your Cross-Cloud Interconnect ports to one of your remote cloud ports.

After Google provisions your connection, we notify you that the connection is ready to use. To begin using the connection, you configure your Google Cloud resources and your remote cloud resources.

For a more detailed overview, see one of the following:

Fixed pricing

Cross-Cloud Interconnect offers fixed port pricing for outbound data transfer for VLAN attachments. This lets you get a fixed monthly invoice for outbound data transfers for the following Cross-Cloud Interconnect connections:

  • Connections with VLAN attachments local to the Google Cloud region where the attachments are connected.
  • Connections with VLAN attachments that are local or remote to the Google Cloud region where the attachments are connected.
  • All connections with VLAN attachments in a geographic location (such as the United States) that are local or remote to the Google Cloud region where the attachments are connected.

A local connection is a VLAN attachment in a metro location with a destination Google Cloud region in the same metro location as your VLAN attachment. For example, if you obtain a VLAN attachment in the Los Angeles, California metro location then the local Google Cloud region is us-west2.

A remote connection is a VLAN attachment in a metro location with a destination Google Cloud region not in the same metro location as your VLAN attachment. For example, if you obtain a VLAN attachment in the Portland, Oregon metro location, then there isn't a local Google Cloud region available within that metro location.

Similarly, if you create a VLAN attachment in the Los Angeles, California metro location, then outbound data transfers from the us-east4 Google Cloud region are considered remote connections, because they are located in different metro locations.

The following Google Cloud regions only offer support for fixed pricing for remote VLAN attachments on Cross-Cloud Interconnect:

  • us-west1
  • us-east1
  • europe-west4
  • europe-north1
  • asia-east1

To request a Cross-Cloud Interconnect connection with fixed port pricing, contact your account team.

Service-level agreement

Cross-Cloud Interconnect uses the Cloud Interconnect service level agreement (SLA). Since Cross-Cloud Interconnect is a shared responsibility model between Google Cloud and another cloud service provider, the Cross-Cloud Interconnect SLA covers Google Cloud's infrastructure up until the point it's transferred to the other cloud service provider.

To qualify for the SLA, your Cross-Cloud Interconnect configuration must use one of the approaches described in the following sections.

Both of these approaches require redundancy. You incorporate redundancy by, at a minimum, locating connections in two edge availability domains. An edge availability domain is a zone within a metropolitan area. Using multiple edge availability domains maximizes availability because two domains in the same metropolitan area are typically not down for maintenance at the same time.

Minimum requirement

The Cloud Interconnect SLA requires you to have, minimally, two connections: a primary connection and a redundant connection, each in a different edge availability domain of a metropolitan area. This approach gives you 99.9% availability.

Figure 1. Set up Cross-Cloud Interconnect with 99.9% availability.
Figure 1. Cross-Cloud Interconnect configuration that has 99.9% availability (click to enlarge).

High availability

For more critical applications, configure two pairs of connections. Each pair must be located in a different metropolitan area. Within each metropolitan area, you must use two different edge availability domains. This approach gives you 99.99% availability.

Figure 2. Set up Cross-Cloud Interconnect with 99.99% availability.
Figure 2. Cross-Cloud Interconnect configuration that has 99.99% availability (click to enlarge).

Limitations

This section describes the limitations of Cross-Cloud Interconnect.

Locations

Connections are supported only in certain locations. For details, see the following documents:

IPv6

IPv6 is supported only when you exchange IPv6 routes over a Border Gateway Protocol (BGP) session that was established between two IPv4 addresses.

What's next