Pre-bursting Cloud Resource Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud bursting strategies fail to reduce costs, do not provide a cost-benefit analysis, and do not account for security and performance requirements when scaling to meet short-term demand, often resulting in inefficient resource allocation and potential service level agreement violations.
Innovation Solution
A method and system that predict and automate the scaling of cloud resources by 'pre-bursting' before actual demand, using a service broker for resource reservations through reverse auctions to form a hybrid cloud, ensuring security and performance requirements are met while minimizing costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If cloud resources are scaled up to meet short-term demand spikes, then service reliability is improved, but cost increases
Solution Approach 1:
The system performs preliminary actions by predicting future demand spikes using historical data and machine learning algorithms before they occur. Resources are pre-provisioned and pre-configured in advance, allowing the system to respond to demand spikes immediately without last-minute scaling operations, thus maintaining service level agreements while optimizing cost through advance planning
Solution Approach 2:
The system continuously monitors cloud resource usage, demand patterns, and service level agreement compliance in real-time. This feedback loop enables dynamic adjustment of resource allocation strategies, allowing the system to learn from past performance and optimize the balance between reliability and cost over time through iterative improvement
2Productivity
If cloud bursting is implemented to address demand spikes, then productivity is improved, but resource allocation efficiency deteriorates
Solution Approach 1:
The system performs preliminary resource provisioning and configuration before demand spikes occur. By predicting future demand using historical data and machine learning, resources are prepared in advance, eliminating the time-consuming process of last-minute resource allocation and enabling immediate response to demand spikes
Solution Approach 2:
The system implements automated resource allocation using machine learning models that independently predict demand patterns and trigger provisioning actions without human intervention. This self-service capability eliminates manual resource allocation processes, significantly reducing the time required to respond to demand changes while maintaining high productivity
3Adaptability or versatility
If external cloud resources are used to meet short-term demand, then adaptability is improved, but security risk increases
Solution Approach 1:
The system introduces an intermediary layer consisting of automated policy enforcement mechanisms and security validation processes between the internal cloud environment and external cloud resources. This intermediary layer continuously monitors and enforces security policies, validating that external resource provisioning meets organizational security requirements before allowing access, thus enabling scalable cloud bursting while maintaining security control
Data Source
AI summary
In a cloud computing environment customers of the cloud believe they have instantaneous access to unlimited resources however to satisfy this with finite resources there are times when resources could have to be acquired from an external cloud with potentially different security capabilities and performance capabilities. A method and system are therefore disclosed to reduce cost incurred while scaling to an external cloud to meet short term demand and to take into account security and performance requirements of customers. The proposed method and system provide automation and prediction capabilities to help with the decision of growing cloud resources or temporarily becoming a hybrid cloud. By “pre-bursting” the cloud in anticipation of a cloud burst the growth in resources can be predicted and performed (with security and load balancing in mind) prior to actual cloud consumer requests.