TL:DR
- Reliable cloud connectivity is as critical as cloud services themselves, as distributed AI workloads demand high-performance, low-latency connections that basic internet access cannot consistently deliver.
- Google’s Verified Peering Provider program simplifies enterprise cloud interconnection by letting trusted partners manage peering relationships, removing the need for specialized in-house networking expertise.
- Equinix’s Gold-tier VPP status & Equinix Fabric access give enterprises flexible, carrier-neutral cloud connectivity options across global metros to support diverse hybrid multicloud workloads.
Today’s enterprises can’t thrive without cloud services. But how they access those services matters. They often need to connect to different cloud environments. In fact, a recent survey of global IT leaders conducted by Foundry and sponsored by Equinix found that 50% of all enterprise workloads are now running in hybrid or multicloud setups.
In many ways, you could say that your connectivity is now your cloud. That’s because choosing the right cloud services won’t do you any good without reliable, high-performance connections to those services.
As enterprise architectures become more distributed, networking is no longer just about reaching the cloud; it increasingly determines how effectively organizations can use the cloud. The quality of the connection influences everything from application performance and user experience to the success of latency-sensitive AI workloads. In fact, at a time when enterprises have dedicated millions to building powerful AI infrastructure, poor network performance could prevent them from realizing return on investment.
Distributed infrastructure is the new standard for AI. In his recent keynote at Google Cloud Next, Bikash Koley, VP of Google Global Infrastructure, argued that “the continent is the new data center.” This means that Google recognizes complex, AI-driven workloads often need to span multiple data center sites. And this only works if the different sites are properly connected.
The advent of distributed AI infrastructure is one of the key factors that’s forcing enterprises to reconsider the way they connect. Businesses that have traditionally used internet-based cloud connectivity may not be able to get the kind of deterministic, synchronized throughput that their distributed AI workloads demand. In addition, they may lack the technical expertise needed to establish secure, private, high-throughput connections that can keep up with the fast-moving AI landscape.
Enterprises need alternatives to peering
Peering is one of the most reliable, lowest-latency ways for two organizations to connect, but it’s also one of the most complex to set up. It requires specialized networking expertise to establish connectivity, plus a formal peering agreement between both parties.
That complexity puts peering out of reach for a lot of organizations. Without the right expertise in house, most default to basic internet connectivity simply because it’s more convenient. That’s why organizations—especially in the AI era—need an alternative: a way to get enterprise-grade connections without the overhead of standing up a full peering partnership.
Increasingly, enterprises are looking for connectivity models that deliver the performance benefits of peering while operating more like a managed service. They want access to AI-ready cloud services that can scale globally without requiring specialized networking expertise in every market.
Google Cloud recognized this need, and responded by redefining how they approach connectivity. The Verified Peering Provider (VPP) program allows enterprises to bypass the complexity of peering, accessing Google-verified connectivity from a roster of trusted partners instead.
Google VPP offers all of the performance with none of the complexity
Google VPP reflects a simple principle: The connection to cloud services matters as much as the services themselves. The program delivers all the benefits of peering without the usual complexity, because Google’s partners manage the entire peering relationship on customers’ behalf.
With VPP, enterprises no longer need to manage complex routing in-house. Setting up reliable, high performance connectivity becomes as straightforward as setting up a basic internet connection.
VPP partners are held to a clear standard. Each one must demonstrate redundant physical connectivity to Google, offer enterprise-class interconnection products, and meet Google’s guidelines for minimizing latency and maintaining resiliency, even during outages. Partners also maintain direct operational contacts with Google for fast troubleshooting when issues arise, and Google periodically re-validates them to confirm they still meet these standards.
Enterprises can choose the partner and networking solution that best fits their needs and gain access to Google Cloud resources worldwide.
This increasingly matters as AI workloads grow more distributed. Whether organizations are running AI powered applications, agentic workflows, or real-time inference, network performance plays a growing role in application responsiveness and user experience. Reliable, low-latency access to cloud services has become a foundational requirement for AI success.
Equinix offers cloud networking choice
Equinix is proud to have achieved Gold-tier status in the Google Gold Verified Peering Provider (VPP) program, meaning that Google has verified our highly redundant connectivity model in 26+ metros across five continents.
We welcome Equinix as a Gold Verified Peering Provider. By significantly expanding our highly available network interconnect, we are ensuring that our mutual customers benefit from the highest standards of connectivity. Equinix’s global footprint, combined with Google’s global network, provides the robust, low-latency foundation enterprises need to succeed in a cloud-first world.”Lars Maartensson, Director of Google Edge Capacity Delivery, Google
Equinix’s ability to offer global carrier-neutral connectivity, colocation and private interconnection makes us unique in the VPP program. Rather than forcing customers into a single connectivity model, Equinix enables them to match connectivity choices to workload requirements. This level of flexibility will be essential for enterprises that need to support many different workloads distributed across many different environments.
Google VPP is accessible via Equinix Internet Access. In addition, access to Google Cloud on-ramps across 48 different sites is available via Equinix Fabric® virtual connections. This means that enterprises can keep their private cloud traffic local to ensure low latency. They also get the reliability and security that comes from knowing sensitive traffic never has to touch the public internet.
Google’s AI-native Cloud Interconnect was designed with traffic differentiation and flexible connection options in mind. As mentioned in a recent Google blog post, those options include both direct fiber peering and colocation facilities like those offered by Equinix. Customers can choose from supported colocation facilities throughout the world to reach Google Cloud Interconnect.
Dojo, a U.K.-based payment technology provider, is one Equinix customer that’s capitalized on our diverse networking solutions. They’re using Equinix Internet Access for internet-based connectivity and Equinix Fabric for private interconnection with multiple cloud providers, including Google Cloud. The company can now process payments 80% faster than the industry average.
Read the Dojo case study to learn more.
Equinix customers can choose a mix of different Google Cloud connectivity options for the following reasons:
Google VPP via Equinix Internet Access:
- Provides simplified access to Google services, with enterprise-class performance and reliability that avoids the public internet
- Supports SASE, SD-WAN and branch traffic
- Prioritizes simplicity and speed of onboarding
Google Cloud Interconnect (GCI) via Equinix Fabric:
- Supports data-center-to-cloud workload connectivity
- Enables hybrid cloud architectures
- Meets the data movement requirements of large-scale AI training workloads
- Prioritizes data privacy and deterministic performance
Organizations typically use Google VPP connectivity via Equinix Internet Access when they need optimized, carrier-neutral access to Google’s public services—including Google Cloud, Google Workspace, and internet-facing applications—without managing a private connection. They use Google Cloud Interconnect via Equinix Fabric when they need a dedicated, private Layer 2 circuit to Google Cloud with defined SLAs and reserved bandwidth—typically for data-intensive or compliance-sensitive workloads.
Since Equinix is a vendor-neutral provider, customers can easily set up their diverse Google Cloud connections alongside connections to multiple other cloud providers, managing all those connections under one roof. They can also connect easily to a variety of AI ecosystem partners, including model providers and neoclouds that offer AI compute infrastructure–such as GPUs, TPUs and other specialized accelerators—as a Service.
As hybrid multicloud architectures continue to evolve, cloud connectivity is becoming a strategic design decision that directly impacts business results, rather than a technical afterthought. By giving enterprises the flexibility to choose the right connectivity model for each workload, Equinix helps them build cloud architectures that are not only connected, but also ready for whatever comes next.
The enterprises that win in the next phase of AI will be the ones that can best eliminate networking friction. This is exactly what connectivity options from Equinix can help them do.
Learn how enterprises throughout the world are tackling hybrid multicloud challenges head-on with help from diverse networking solutions: Read the research paper, “Connectivity as the competitive edge: The state of hybrid multicloud networking.”
Subscribe to the Equinix Blog