AI Container Autoscaling for 5G Core Traffic Spikes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud-native 5G core networks face inefficiencies in autoscaling due to sub-optimal resource utilization and inability to adapt quickly to changing network demands, leading to high latency and resource waste.
Innovation Solution
Implementing a machine learning model to proactively make autoscaling decisions based on specific network functions and operational requirements, using historical and real-time metrics to predict future states and optimize container scaling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional cloud-native autoscaling is used in 5G core networks, then network resources can be managed automatically, but the system exhibits high latency and sub-optimal resource utilization when adapting to sudden traffic changes
Solution Approach 1:
The patent applies preliminary action by using machine learning models to predict future network traffic patterns and proactively scaling containers before actual traffic spikes occur. The system continuously trains ML models on historical traffic data to forecast future states, enabling preemptive autoscaling decisions that reduce latency and improve responsiveness to traffic changes.
2Productivity
If traditional autoscaling methods are used, then container scaling can be automated, but resource waste occurs due to inability to accurately predict future network demands
Solution Approach 1:
The system uses machine learning models to predict future network traffic patterns and proactively scales containers before actual demand occurs. This preliminary prediction and action approach allows the system to allocate resources more accurately, avoiding both over-provisioning and under-provisioning, thereby reducing resource waste while improving utilization efficiency.
Solution Approach 2:
The patent implements feedback mechanisms where the system continuously monitors actual network traffic and compares it against ML predictions. This feedback loop allows the system to learn from actual performance, refine its predictions, and adjust scaling decisions accordingly, optimizing resource allocation over time and reducing waste from mismatched resource provisioning.
Data Source
AI summary
The present disclosure relates to a communication method and system for converging a 5th-Generation (5G) communication system for supporting higher data rates beyond a 4th-Generation (4G) system with a technology for Internet of Things (IoT). The present disclosure may be applied to intelligent services based on the 5G communication technology and the IoT-related technology, such as smart home, smart building, smart city, smart car, connected car, health care, digital education, smart retail, security and safety services. A method performed by an artificial intelligence (AI) module for autoscaling containers of a cloud-native core network with containerised network functions is provided. The method comprising requesting, from at least one metrics server, at least one metric required to make an autoscaling decision with respect to at least one set of containers; receiving the at least one metric from the at least one metrics server; processing the received at least one metric, using a trained machine learning (ML) model, to make an autoscaling decision with respect to the set of containers; and implementing the autoscaling decision with respect to the set of containers.