Private 5G Network (P5G) with Google Cloud Platform

Introduction

With technology evolution to 5G, enterprises & industries are going through transformation for better efficiencies and new revenue streams. With 5G, wireless networks have changed drastically especially with user expectations for immersive experiences (AR/VR/4K Streaming), reliability, low latency, high throughput and high volume of connected devices. Considering this, existing wireless networks be it WiFi or LTE are not enough and Private 5G is ready to meet those user expectations.

5G brings eMBB, URLLC, mMTC, security and energy efficiency as key functional drivers to industrial advancement with following use cases.

  • Manufacturing Industry and Healthcare can use faster and effective inspections through predictive intelligence provided by 5G
  • Oil & Gas, Sea Ports can utilize 5G for Workplace and Worker safety
  • Supply chain, Logistics and Retail operational Efficiency improves with 5G

Private 5G Model

There are two fundamental models for Private 5G network deployments.

  • Enterprises owned both spectrum and Private 5G networks,
  • Operator owned spectrum and provide Private 5G networks as managed service

GCP Private 5G Solution

This year, Google announced private networks solution  “Google Distributed Cloud Edge – (GDCE)” for Telcos and Enterprises. The solution has been built with Cloud Native Platforms.

Google Distributed Cloud Edge 

  • Google Distributed Cloud Edge (GDCE) supports 5G RAN/Core network functions and
  • GDCE allows network functions to be hosted in distributed manner with flexibility of being at Telco edge, Google edge or Enterprise edge locations depending on the use case
  • GDCE is a fully managed, on-prem infrastructure solution providing Hardware, Google Container optimized OS and Container Management/Orchestration for localized workloads
  • GDCE is powered by Anthos, an hybrid and multi-cloud software platform for distributed cloud
  • Anthos as software layer and GDCE as a managed infrastructure platform can be managed & exposed via APIs providing a cloud like experience of your on-prem remote edge compute environments wherein logging, the monitoring, the provisioning everything is exposed to cloud console

Architecture & Deployment Models

Following figure shows the high level architecture is using Centralized Google Cloud 1P services (in blue boxes) with common control and management along with hosting of various 5G RAN and Core control plane functions (in green boxes) for Data, Voice and SIM management services. The Data plane is distributed  to on-prem Enterprise A location hosting 5G RAN and Core User plane functions along with Multiservice Edge Computing (MEC) applications for latency sensitive use cases.

Image Source: google

The on-prem edge locations are connected back to Google Cloud public region using Interconnect services like (Dedicated Interconnect/Partner Interconnect/Cloud VPN) while endpoints (IoT devices, Mobile devices, Compute devices) are connected over 5G radio network using 5G-NR (new radio) in an enterprise environment.

In the case of Enterprise B, the architecture provides flexibility of hosting even the 5G RAN and Core User plane functions along with MEC applications in GCP public region enabled by GKE/GCE service serving Private 5G use cases those are not latency sensitive.

Deployment Model : Enterprise Edge

This deployment model is best suited for either Enterprise owned Private 5G — Standalone Non-Public Network (SNPN) or Telco owned spectrum and offers Private 5G networks as managed service enabling the use cases that require on-premise ultra reliability, low latency and mobility.

Image Source: google

Key attributes of the deployment model with GDCE deployed at Enterprise edge locations are:

  • A converged platform for Private 5G Network functions (nearest to 5G-NR radio) along with Enterprise applications like Computer Vision, Predictive maintenance, Loss prevention etc.
  • Empowering ultra low latency use cases applicable across various industry verticals like retail, manufacturing, mining etc.
  • All the distributed GDCE locations are still managed centrally providing many Google Cloud capabilities like Common Control and Management, Analytics, AI/ML and CI/CD operations for distributed edge locations
  • Secured connectivity with all the traffic between GDCE and Google Cloud over SD-WAN or Internet is encrypted
  • The MEC/Enterprise edge applications are not limited to the above model but can be any cloud-native applications built for specific industries either by ISVs or by Enterprises themselves as per the use case requirements
  • This deployment model is best suited for <1ms latency requirements at the 5G radio layer since 5G core functions and MEC applications are hosted next to each other.

Deployment Model : Telco Edge

This deployment model is best suited for Telco owned spectrum and offers Private 5G networks as managed service enabling the use cases that require mobility and are more tolerant to latency requirements.

Image Source: google

Key attributes of the deployment model with GDCE deployed at Telco Network edge locations are:

  • A converged platform for P5G Network functions along with Enterprise applications like ADAS, Telepath, AR/VR and IoT etc.
  • Empowering more latency tolerant use cases applicable across various industry verticals like Smart Cities, Venues connectivity, Healthcare, Entertainment & Gaming.
  • All the distributed GDCE locations at Telco Network edge are still managed centrally providing many Google Cloud capabilities like Common Control and Management, Analytics, AI/ML and CI/CD operations for distributed edge locations
  • Secured connectivity with all the traffic between GDCE and Google Cloud over Dedicated/Partner Interconnect or Internet is encrypted
  • The MEC/Enterprise edge applications are not limited to the above model but can be any cloud-native applications built for specific industries either by ISVs or by Enterprises themselves as per the use case requirements
  • This deployment model is best suited for up to 5ms latency at the 5G radio layer since 5G core functions and MEC applications are hosted next to each other at network edge away from Enterprise locations.
  • The Enterprise locations are connected over the 5G network but still maintaining required bandwidth and latency constraints enabled by Network slices for each of the enterprises by the Telco.

Conclusion

Google Cloud Private 5G solution powered by Google Distributed Cloud Edge (GDCE), Partner (Telcos and ISVs) ecosystem and 1P Google assets/capabilities coupled with centrally managed (GCP region) edge infrastructure provides horizontal enabler layer delivering multitude of the use cases across many industry verticals.

It provides open, flexible architecture and deployment models for Enterprises and/or Telcos with future evolution to many new use cases enabled by 3GPP Release 17 and Release 18 features/functionality for Private 5G (P5G) networks.

References

Related Posts

You may also like...