Predictive Workload Elasticity for Edge Network Function Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to efficiently manage and scale network functions across edge sites in a telecommunications network, particularly in scenarios involving service chaining and sudden load changes, leading to potential service degradation or denial of service.
Innovation Solution
Implementing a predictive model that forecasts network function usage spikes and dynamically adjusts instance allocation across edge sites and data centers, considering factors like latency, computation budgets, and resource availability, while maintaining seamless user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If network function instances are scaled up to meet user demands, then service availability and performance are improved, but resource cost and complexity increase
Solution Approach 1:
The system implements self-service through automated orchestration that monitors workload demands and automatically scales network function instances without human intervention. The orchestration system detects when service levels are approaching thresholds and autonomously provisions additional instances or redistributes workloads, eliminating the need for manual management while maintaining high service availability.
Solution Approach 2:
The patent employs feedback mechanisms where the orchestration system continuously monitors service performance metrics and workload patterns. Based on this feedback, the system dynamically adjusts instance allocation - scaling up when service demands increase and scaling down when demands decrease, thereby maintaining optimal service availability while controlling resource costs and management complexity.
2Loss of energy
If network function instances are scaled down to reduce cost, then resource efficiency improves, but service performance and availability deteriorate
Solution Approach 1:
The system applies dynamics by implementing flexible, adaptive instance scaling that responds to real-time workload conditions. Rather than maintaining fixed instance counts, the orchestration system dynamically adjusts the number and distribution of network function instances based on current service demands, ensuring optimal resource utilization while preventing service performance degradation.
Solution Approach 2:
The patent implements preliminary action through predictive scaling mechanisms that anticipate future workload demands based on historical patterns and current trends. The orchestration system provisions instances before peak demands occur, ensuring service performance is maintained during anticipated high-load periods while avoiding unnecessary instance proliferation during low-demand periods, thus optimizing resource costs.
3Reliability
If workload is distributed across multiple edge sites, then service resilience and availability improve, but orchestration complexity and communication overhead increase
Solution Approach 1:
The system applies segmentation by dividing the orchestration functionality into distributed components at each edge site while maintaining centralized coordination for global optimization. Each edge site's orchestration component independently manages local instance provisioning and workload distribution, reducing communication overhead and simplifying local decision-making, while still achieving service resilience through geographic distribution across multiple sites.
4Reliability
If instances are rapidly scaled to meet sudden load changes, then service availability is maintained, but system stability and performance consistency worsen
Solution Approach 1:
The patent implements preliminary action through predictive analytics that forecast upcoming workload surges based on historical patterns and current trends. The orchestration system proactively provisions additional network function instances before the predicted load spike occurs, allowing the system to smoothly handle sudden demand changes without abrupt scaling actions that would destabilize the system, thereby maintaining both service availability and system stability.
Data Source
AI summary
Computing resources are managed in a computing environment comprising a computing service provider and an edge computing network. The edge computing network comprises computing and storage devices configured to extend computing resources of the computing service provider to remote users of the computing service provider. The edge computing network collects capacity and usage data for computing and network resources at the edge computing network. A predictive function is applied to the data to determine a predicted demand on the computing and network resources at a future time interval. Based on the predicted demand, a distribution of workloads is determined.