Predictive Resource Provisioning in Cloud Infrastructure

Solution Overview

Problem

Cloud service providers face inefficiencies in provisioning resources due to unstructured processes, leading to higher costs from unused resources, as they often over-provision to account for peak demands, resulting in mismatched consumer needs.

Innovation Solution

Implementing a predictive resource consumption system that collects and analyzes enterprise data to forecast resource demands, providing a schedule to the cloud infrastructure for precise provisioning of storage and computational resources based on expected events, with feedback loops to update predictions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If additional hardware is provisioned to account for peak demands, then service reliability is improved, but infrastructure cost increases due to unused resources

Engineering Contradiction:

Improveservice reliabilityVSAvoidinfrastructure cost

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The system performs preliminary analysis of historical event data and enterprise operations to predict future resource consumption patterns. By proactively identifying upcoming resource demands before they occur, the system enables advance resource provisioning that matches actual usage patterns, avoiding both over-provisioning and under-provisioning scenarios

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements continuous feedback loops where actual resource consumption data is collected, compared against predictions, and used to refine future predictions. This feedback mechanism enables dynamic adjustment of resource provisioning strategies, improving accuracy over time and optimizing the balance between reliability and cost

Inventive Principle:

Principle #23Feedback

2Speed

If unstructured provisioning is used to respond to real-time demand, then responsiveness is improved, but resource allocation accuracy deteriorates leading to over-provisioning

Engineering Contradiction:

ImproveresponsivenessVSAvoidresource allocation accuracy

Core Design Contradiction:

SpeedVSMeasurement precision

Solution Approach 1:

The system performs preliminary analysis of historical event data and enterprise operations to predict future resource consumption patterns. By proactively identifying upcoming resource demands before they occur, the system enables advance resource provisioning that matches actual usage patterns, avoiding both over-provisioning and under-provisioning scenarios

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system transitions from static, rule-based provisioning to dynamic, data-driven provisioning. Resource allocation continuously adapts based on real-time consumption patterns, historical trends, and predicted future demands, enabling the system to respond optimally to changing conditions while maintaining high accuracy

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS9606840B2Enterprise data-driven system for predictive resource provisioning in cloud environments

Publication Date: 2017.03.28 SAP SE

AI summary

Implementations of the present disclosure include methods, systems, and computer-readable storage mediums for predicting resource consumption in cloud infrastructures. Implementations include actions of receiving event data from one or more enterprise data sources, determining that an event associated with the event data is a known event, retrieving resource consumption information associated with the event, and providing a resource consumption schedule to a cloud infrastructure, the resource consumption schedule indicating resources expected to be consumed during execution of the event.