AI Resource Estimation for Mobile Packet Core Cloud Deployments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual network sizing for wireless networks in hybrid or public cloud deployments is labor-intensive and often does not yield optimized resource estimates, leading to inefficient resource allocation and increased costs.
Innovation Solution
An AI-based optimization model that uses call models and traffic parameters to dynamically calculate and adjust resource allocation based on current network traffic patterns, incorporating machine learning for seasonal trends and historical data to optimize resource scheduling and cost estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual network sizing is performed by engineers, then resource estimation can be obtained, but the process becomes labor-intensive and time-consuming
Solution Approach 1:
The patent replaces the manual mechanical process of engineer-based resource sizing with an automated machine learning system. The ML model automatically processes network traffic data, call models, and service specifications to generate resource estimates, eliminating the need for manual engineering analysis while maintaining or improving accuracy.
Solution Approach 2:
The system enables self-service resource estimation by automatically gathering input parameters, processing them through the ML model, and generating deployment recommendations without requiring manual intervention from engineers. The system serves itself by continuously learning from historical data and improving its estimation capabilities.
2Quantity of substance
If manual resource sizing is performed, then deployment cost estimation can be obtained, but the estimates are not optimized and labor costs increase
Solution Approach 1:
The patent implements dynamic resource allocation by using the ML model to continuously adapt resource estimates based on current network conditions, traffic patterns, and service requirements. This dynamic approach replaces static manual sizing, allowing the system to optimize resource allocation in real-time and improve productivity by allocating the right amount of resources to the right services.
Solution Approach 2:
The system incorporates feedback mechanisms where the ML model continuously learns from actual network performance data, deployment outcomes, and cost information. This feedback loop enables the system to refine its resource estimation accuracy over time, optimizing resource allocation and reducing waste while improving productivity.
3Reliability
If resources are allocated based on peak traffic analysis, then network reliability is maintained, but resource utilization becomes inefficient during non-peak periods
Solution Approach 1:
The patent applies dynamic resource allocation that adjusts resource provisioning based on predicted traffic patterns rather than static peak-based allocation. The ML model forecasts future network conditions and dynamically scales resources accordingly, maintaining reliability during peak periods while reducing resource consumption during low-traffic periods, thus improving overall utilization efficiency.
Solution Approach 2:
The system performs preliminary actions by using the ML model to predict future network traffic and proactively allocate resources before peak demand occurs. This predictive capability allows the system to prepare appropriate resource levels in advance, ensuring reliability when needed while avoiding over-provisioning during off-peak periods, thereby reducing energy waste.
Data Source
AI summary
Computing and network capacity are allocated in a computing environment provided by a virtualized computing service provider. An AI-based optimization model is run to quantify current network traffic in the computing environment based on processing and storage usage patterns using key performance indicators (KPIs). The quantified current network traffic is used to calculate, by a sizing and capacity model of the AI-based optimization model, a number, types, and sizes of disk storage and processing resources based on estimated cost.