Ericsson Partners with Google Cloud to Deliver On-Demand 5G Core via SaaS Model

Ericsson Partners with Google Cloud to Deliver On-Demand 5G Core via SaaS Model

Ericsson has integrated its virtualized 5G core with Google Cloud’s public infrastructure, enabling communications service providers (CSPs) to deploy mobile core networks globally on a pay-as-you-go basis. This partnership leverages Google Kubernetes Engine (GKE) to run Ericsson’s cloud-native control and user plane functions, designed to support ultra-low latency, high-availability requirements in telecom networks. 

“We’re not just talking about the control plane — this includes the user plane too,” said Eric Parsons, VP and Head of Emerging Segments, Cloud Software & Services at Ericsson. He emphasized the platform’s ability to meet sub-second convergence demands, critical for mobile networks sensitive to traffic interruptions. 

Ericsson’s solution is offered as a Software-as-a-Service (SaaS) package, fully managed and maintained by Ericsson. It enables CSPs to launch core networks within minutes, according to Muninder Singh Sambi, VP & GM of Networking and Security at Google Cloud. The architecture allows for instant scalability and cost-efficiency, where operators only pay for what they use. 

Native Cloud Integration vs Traditional Virtualization 

While Ericsson has previously delivered its virtualized core for on-premises environments using VMware and Red Hat, this latest integration is cloud-native and optimized specifically for Google Cloud. Parsons explained that this move goes beyond a typical multicloud approach and instead leverages the full elasticity and infrastructure scale of the public cloud. 

“This is not just lifting and shifting the core logic,” Parsons said. “We’re deeply investing in the networking layer to make it native to Google Cloud.” This new Ericsson On-Demand service builds on earlier collaborations between Ericsson and Google Cloud, including 5G RAN integration and an API-focused joint venture aimed at unlocking new monetization paths for telecom operators. 

Target Audience: CSPs of All Sizes 

Ericsson’s SaaS 5G core is primarily targeted at CSPs — both large operators looking to expand their global footprint and simplify operations, and smaller or greenfield operators beginning their 5G journey. The offering enables quick deployment in new regions without the need for extensive physical infrastructure. 

That said, not all CSPs are shifting to public cloud just yet. Many large telcos continue to prefer private, in-house cloud environments. For example, Verizon’s SVP of Technology Planning, Adam Koeppe, confirmed their 5G core runs on Verizon-owned infrastructure to maintain end-to-end control over the customer experience. AT&T’s Igal Elbaz echoed similar sentiments, stating their 5G core operates entirely within AT&T data centers. However, both executives acknowledged that public cloud will play a growing role, especially in enterprise use cases or hybrid architectures, where integration with on-prem assets is necessary. 

Elastic Augmentation & Disaster Recovery 

Ericsson believes public cloud-based cores can complement existing private networks by providing on-demand burst capacity, particularly useful during traffic surges or for disaster recovery (DR) purposes. “Instead of overbuilding data center capacity, CSPs can scale instantly on Google Cloud,” Parsons added. 

Google Cloud’s Telecom Strategy 

Google Cloud’s approach has been to adapt to each telco’s cloud journey, said Sambi. Whether CSPs want a fully managed 5G core or hybrid integrations with existing infrastructure, the goal is to offer flexible, modular solutions. 

This philosophy is already reflected in Google Cloud’s partnerships with Bell Canada, Verizon, and Telus, among others. “We’re not here to define the path,” Sambi said. “We’re here to support their transformation—whether they want full SaaS, hybrid, or burst models.” 

Market Outlook 

Google Cloud’s push into telecom is part of a broader industry trend. AWS has secured partnerships with operators like Comcast, Boost Mobile, and O2 Telefónica, while Microsoft has taken a more focused approach in supporting select telecom initiatives. 

According to Dell’Oro Group, hyperscalers like AWS, Azure, and Google Cloud are expected to host 9% of global 5G standalone (SA) workloads by 2027, underscoring the growing role of cloud providers in future mobile infrastructure. 

Leave a Reply

Your email address will not be published. Required fields are marked *

Reliance Jio's ₹799 Plan with 1.5GB Daily Data is Ideal for Most Indian Users
5G/6G

Reliance Jio’s ₹799 Plan with 1.5GB Daily Data is Ideal for Most Indian Users

Summary: Reliance Jio’s ₹799 prepaid plan is a value-focused option for users seeking 84 days of validity with 1.5GB of daily data, unlimited voice calls, 100 SMS per day, and complimentary access to JioTV and JioAICloud. Designed for customers who prioritize essential connectivity over OTT subscriptions or unlimited 5G, the plan offers a balanced combination […]

Read More
Airtel Clarifies It Has Never Restricted Hotspot Usage Under Its Unlimited 5G Data Plans
5G/6G

Airtel Clarifies It Has Never Restricted Hotspot Usage Under Its Unlimited 5G Data Plans

Summary: The controversy surrounding Airtel’s unlimited 5G hotspot policy was largely driven by a misunderstanding of its terms and conditions rather than any actual technical restriction. Although a clause addressed hotspot usage to prevent commercial misuse, Airtel never disabled the hotspot feature for regular users, who continued sharing data with personal devices without interruption. With […]

Read More
India's 5G Infrastructure Faces Challenges in Supporting AI Workloads
5G/6G

India’s 5G Infrastructure Faces Challenges in Supporting AI Workloads

Summary: India’s 5G network has expanded quickly, led by Reliance Jio, Bharti Airtel, and the ongoing rollout by Vodafone Idea, placing the country ninth globally for median 5G download speeds. However, an Ookla report highlights that relatively low upload speeds and average multi-server latency of 51.6 ms could limit the performance of AI-powered applications, particularly […]

Read More
Copyright @ 2025 Bharatnet. All rights reserved.