MWC26: Google Cloud tackles telco data fragmentation

• Cloud Spanner Graph acts as a digital twin offering a live network map to help tackle problems
• Google Cloud is also looking to take on siloed data to help give AI greater context in network operations

BARCELONA – #MWC26 – Google Cloud has unveiled new solutions aimed at helping telecom operators manage their data resource and adopt agentic AI.

The hyperscaler warned that most telcos are still navigating “massive amounts of fragmented information trapped in disconnected silos” that it calls “data swamps”. 

To help drag telcos out of those swamps and put that data to good use, it is leveraging its data warehouse system, BigQuery, to help operators prepare their datasets by integrating its AI platform, Gemini, directly into data lifecycles.

The hyperscaler’s Cloud Spanner Graph is also playing a role by acting as a digital twin and giving AI the necessary context by providing a live, temporal map that unifies relationships across the network.

And in a related move, the tech giant said it has partnered with data management specialist DigitalRoute to launch reusable data pipelines that act as high-speed filtration systems to transform “chaotic network signals into a single source of truth”. 

Google Cloud is also addressing challenges in telco core platform operations with a bundle of tools – BigQuery, Spanner and Vertex AI to provide an AI foundation and help unify data, and Gemini Enterprise as an agentic platform. According to the hyperscaler, this combination avoids the pain of manual data sync processes between departments by enabling agents to engage with OSS and BSS systems and identify the correlation between technical network events and customer outcomes. 

The hyperscaler is also aiming to help operators “bridge the gap between 5G investment and revenue” using its evolved Autonomous Network Operations framework, which it believes can help network operators reach level 4 or even level 5 on the TM Forum’s Autonomous Networks (AN) framework (measured in levels 0 to 5). 

This latest iteration of the framework, first unveiled last year, includes Cloud Spanner Graph which, with its digital twin capabilities, allows AI agents to “see” the state of the network and query historical data to carry out root-cause analysis. Working with network intelligence partner NetAI, Google Cloud is helping Spanish operator MásOrange to use these models to resolve incidents with machine-speed accuracy.

It is also training graph neural networks in Vertex AI to not just monitor the network but predict problems too.

According to Google, Vodafone is using the Autonomous Network Operations framework to allow its network to automatically fix broken connections, while it is working with Deutsche Telekom on MINDR (Multi-Agentic Intelligent Network Diagnostics & Remediation), which uses multiple agents to resolve issues across different network domains simultaneously. 

- James Pearce, Editor, TelecomTV