Google has announced the general availability of Cloud WAN, a new managed wide area network (WAN) solution that leverages Google’s global network infrastructure.
The company states that Cloud WAN is designed to provide a secure, reliable, high-performance enterprise backbone, offering an alternative to traditional multiprotocol label switching (MPLS) networks and complex SD-WAN deployments.
Google’s global network comprises 202 points of presence (PoPs), over 2 million miles of fiber, and 33 subsea cables, backed by a 99.99% reliability SLA. The company says this infrastructure powers both Google Cloud and its Cross-Cloud Network solutions, and now forms the foundation for Cloud WAN.
The announcement addresses the increasing complexity of enterprise networking, driven by the adoption of SaaS and cloud applications, and the rise of AI. Google notes that traditional MPLS networks are high-cost, while SD-WAN with direct internet access (DIA) introduces application performance and security challenges. The emergence of AI, with its distributed infrastructure and demands for scalability, security, and cost-effectiveness, further complicates network requirements.
In a Google Cloud blog post, Subhasree Mandal, a lead on the Global Network Technology team, explained that the AI era brought new challenges, starting with the scale of traffic AI-powered apps and model training sent to Google’s network:
We introduced a multi-shard horizontal network architecture to swiftly grow capacity. Here, each shard is essentially a different instance of the network that exists independently. We can scale the network within each shard and increase the number of shards as demand increases. It’s like we’re offering capacity from multiple ISPs, which ensures redundancy, too.
Cloud WAN aims to simplify enterprise connectivity by offering a unified solution for connecting geographically dispersed data centers, branch offices, and campuses.
(Source: Google Cloud website)
In a blog post, the company highlights two primary use cases for Cloud WAN:
- High-performance, cross-region connectivity: For large, global organizations with extensive data center networks, Cloud WAN offers flexible connectivity options, including Cloud Interconnect and Cross-Cloud Interconnect (for multi-cloud connectivity). A new Cross-Site Interconnect feature, currently in preview, provides layer two private connectivity between data centers.
- Migrating branch and campus networks: Cloud WAN extends Google’s Premium Tier network to securely connect branch offices and campuses to cloud resources and SaaS applications. This provides a managed solution with integrated security and a lower total cost of ownership (TCO). Google’s Premium Tier network is designed to optimize application performance by routing traffic through Google’s network as efficiently as possible, with extensive peering connections.
(Source: Google blog post)
Matteo Di Maggio, Senior Platform Group Manager, Connectivity and Voice at Nestle, concluded in a YouTube video on Cloud WAN:
The transition to Cloud WAN has already shown performance improvements and promises further cost savings by replacing older technologies.
Google claims that Cloud WAN can provide up to 40% faster performance compared to the public internet, and up to 40% savings in TCO compared to customer-managed WAN solutions. The TCO reduction is attributed to reduced reliance on MPLS, consolidation of carrier-neutral facility deployments, and flexible pricing options.
Google also highlights that enterprises can work with their existing managed service providers (MSPs) to migrate and operate Cloud WAN and that global system integrators (GSIs) like Accenture, HCLTech, and Wipro offer Cloud WAN services.
