Cloud Native RAN Workload Scaling via Dynamic Container Adjustment

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional radio access networks (RANs) struggle with dynamically scaling subscriber capacity and processing resources to match varying demand, leading to underutilization of compute resources during off-peak periods.

Innovation Solution

Implementing a cloud native RAN system that uses cloud technologies to dynamically scale processing capacity by monitoring and adjusting resource assignments in containers within a Kubernetes computing environment, allowing for automatic scaling out and scaling in of subscriber handling pods.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional radio access networks are configured for peak subscriber processing capacity demands, then they can handle maximum load, but compute resources become under-utilized when processing capacity falls below peak demand

Engineering Contradiction:

Improveability to handle peak subscriber demandVSAvoidunder-utilization of compute resources

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The patent implements dynamic scaling of cloud native RAN workloads by monitoring subscriber demand and automatically adjusting the number of container instances and processing resources. The system transitions from static peak-based configuration to dynamic demand-based configuration, allowing resources to scale up during peak periods and scale down during off-peak periods, thereby resolving the contradiction between handling peak demand and avoiding resource under-utilization

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes key parameters including the number of container instances, CPU allocation, and memory allocation based on real-time subscriber demand metrics. By dynamically adjusting these parameters rather than maintaining fixed peak-level allocations, the system achieves both reliable peak handling and efficient off-peak resource utilization

Inventive Principle:

Principle #35Parameter changes

2Loss of energy

If cloud native RAN workloads are scaled dynamically to match subscriber demand, then resource utilization improves, but service interruptions may occur during scaling operations

Engineering Contradiction:

Improveresource utilization efficiencyVSAvoidservice continuity during scaling

Core Design Contradiction:

Loss of energyVSReliability

Solution Approach 1:

The system performs preliminary actions by pre-warming container instances before actual demand increases and by preparing standby resources in advance. This allows smooth transitions during scaling operations without service interruptions, as resources are ready before needed rather than being created on-demand during critical periods

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements cushioning mechanisms by maintaining buffer resources and using gradual scaling approaches. During scaling operations, the system keeps spare capacity available and transitions resources incrementally rather than all-at-once, providing a cushion that prevents service interruptions even if scaling operations encounter issues

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

3Productivity

If upgrading cloud native RAN clusters is performed to improve performance, then system capabilities increase, but service interruptions occur during the upgrade process

Engineering Contradiction:

Improvesystem performance and capabilitiesVSAvoidservice continuity during upgrades

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent segments the RAN cluster into multiple independent container instances that can be upgraded separately. Rather than upgrading the entire cluster at once, the system divides workloads across multiple segments and performs incremental upgrades on individual segments, allowing continuous service delivery while improving system capabilities

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system maintains continuous useful action during upgrades by ensuring that at least one container instance remains operational throughout the upgrade process. The upgrading mechanism preserves service continuity by transitioning from old to new versions without complete service interruption, thereby achieving performance improvements while maintaining reliability

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS12335787B2Scaling of cloud native radio access network workloads in a cloud computing environment

Publication Date: 2025.06.17 RAKUTEN SYMPHONY INC

AI summary

A method, an apparatus, and a computer program product for scaling one or more processing resources in a wireless communication system. One or more processing resources being assigned to one or more containers in a plurality of containers of a cloud native radio access network for providing communication to at least one user equipment in a plurality of user equipments are monitored. Based on the monitoring, a determination of whether to change an assignment of one or more processing resources in the plurality of containers is made. Based on the determination, the assignment of one or more processing resources is changed.