Predictive AI Cloud Service Turn-Up Automation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud service provisioning methods require significant time and costs for on-demand access, involving manual configuration and inefficient resource allocation, which hinders quick and cost-effective service turn-up for cloud service subscribers.
Innovation Solution
A predictive AI system that uses an AI pipeline and service orchestration server to automatically turn up cloud services based on customer usage data, predicting service needs and provisioning resources accordingly, thereby reducing manual intervention and optimizing resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual configuration and conventional provisioning methods are used, then service provisioning can be completed, but significant time and costs are required, reducing productivity
Solution Approach 1:
The AI pipeline performs preliminary actions by analyzing customer usage data and predicting future service needs before actual requests are made. The system provisions cloud services in advance based on predicted demand patterns, eliminating the need for time-consuming manual configuration when services are actually needed.
Solution Approach 2:
The system enables self-service automation where the AI pipeline independently analyzes usage data, predicts service requirements, and triggers provisioning workflows without human intervention. The service orchestration server automatically executes turn-up procedures based on AI predictions, eliminating manual configuration steps.
2Adaptability or versatility
If on-demand cloud service access is allowed, then customer flexibility is improved, but resource allocation efficiency deteriorates due to manual configuration requirements
Solution Approach 1:
The system implements feedback loops where the AI pipeline continuously monitors customer usage data and service provisioning outcomes. This feedback enables the system to learn from actual usage patterns and improve future predictions, allowing flexible on-demand access while optimizing resource allocation through data-driven decisions.
Solution Approach 2:
The system dynamically changes provisioning parameters based on analyzed usage data and predicted demand. Instead of static resource allocation, the AI pipeline adjusts service provisioning parameters in real-time according to actual customer needs, improving both flexibility and resource efficiency.
3Ease of operation
If manual configuration is performed for each cloud service request, then service customization is achieved, but device complexity and operational difficulty increase
Solution Approach 1:
The system performs self-service configuration where the AI pipeline automatically analyzes usage data and determines optimal service provisioning parameters without requiring manual configuration. The service orchestration server executes automated provisioning workflows that simplify the process while maintaining service customization based on predicted needs.
4Productivity
If cloud services are provisioned on-demand, then customer service access is improved, but costs increase due to inefficient resource allocation
Solution Approach 1:
The AI pipeline performs preliminary analysis of usage data to predict service needs before they occur. By provisioning services in advance based on accurate predictions, the system avoids both the cost of maintaining idle standby resources and the cost of emergency provisioning, optimizing resource allocation while ensuring fast service access.
Data Source
Figure 1A
Figure 1B
Figure 2A
AI summary
Novel tools and techniques for predictive AI automated cloud service turn-up are provided. A system includes an AI pipeline and service orchestration server coupled to the Ai pipeline. The AI pipeline includes a processor and non-transitory computer readable media comprising instructions executable by the processor to obtain customer usage data associated with a first customer from one or more customer data sources, wherein the customer usage data is indicative of usage patterns of one or more cloud services by the first customer, and generate, via a predictive model, predicted usage data based on the customer usage data, wherein the predicted usage data includes a prediction of an individual cloud service of the one or more cloud services predicted to be used by the first customer. The service orchestration server may be configured to turn-up the individual cloud service based on the predicted usage data.