CU-UP Auto-Scaling for Dynamic Network Load Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing CU-UP management in telecommunication networks is static and inefficient, leading to suboptimal resource allocation and manual, passive scaling that fails to adapt to dynamic network load and traffic conditions, resulting in poor user experience and resource wastage.
Innovation Solution
A system and method for automatically and dynamically managing the user plane of a network by continuously monitoring load and resource utilization, adjusting the scale of CU-UP through pod and container management in real-time, using an orchestrator system to optimize resource allocation based on predefined thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If networking resources are allocated or reserved according to initial scaling of CU-UP, then optimal CU-UP performance is achieved initially, but the allocation becomes suboptimal when network load varies over time
Solution Approach 1:
The patent implements dynamic scaling of CU-UP by continuously monitoring network load metrics (data throughput, latency, packet loss) and automatically adjusting the number of CU-UP pods and their resource allocation. This transforms the static resource allocation into a dynamic system that adapts to changing network conditions in real-time, resolving the contradiction between initial performance optimization and adaptability to load variations.
Solution Approach 2:
The system establishes a feedback loop where network load performance metrics are continuously monitored and fed back to the scaling manager. Based on this feedback, the system automatically adjusts CU-UP scaling decisions, enabling adaptive resource allocation that maintains optimal performance under varying network conditions.
2Adaptability or versatility
If manual scaling of CU-UP is performed periodically, then resource allocation can be adjusted to network conditions, but the process is inefficient and cannot respond to immediate scaling requirements
Solution Approach 1:
The patent implements self-service automation where the scaling manager autonomously monitors network load, evaluates scaling criteria, and executes CU-UP scaling operations without manual intervention. The system automatically adjusts the number of pods and resource allocation based on real-time network conditions, eliminating the inefficiency of manual periodic scaling and enabling immediate response to scaling requirements.
Solution Approach 2:
The system performs preliminary scaling actions proactively based on predicted network conditions and predefined scaling policies. By monitoring trends and anticipating future load changes, the system can pre-scale resources before performance degradation occurs, improving both adaptability and response efficiency.
3Reliability
If a higher number of pods are allocated to CU-UP, then network performance is improved, but networking resources are wasted when network load is low
Solution Approach 1:
The patent implements dynamic resource allocation where the number of CU-UP pods and their networking resources (CPU, memory, bandwidth) are continuously adjusted based on real-time network load monitoring. When load is high, more pods are allocated to maintain performance; when load is low, pods are reduced to eliminate resource wastage, thus dynamically balancing performance and resource efficiency.
Solution Approach 2:
The system changes key parameters including the number of pods, CPU allocation, memory allocation, and bandwidth allocation based on network load conditions. These parameter adjustments enable the system to optimize the trade-off between network performance and resource consumption, allocating resources proportionally to actual needs rather than maintaining fixed high allocations.
Data Source
AI summary
Provided are system, method, and device for managing capacity thresholds in a network. According to embodiments, the system may include: a memory storage storing computer-executable instructions; and at least one processor communicatively coupled to the memory storage, wherein the at least one processor may be configured to execute the instructions to: obtain information associated with load of a network; and manage, based on the obtained information, user plane of the network.