Cloud Resource Exchange Management via Credit-Based Workload Shifting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud service providers face limitations in managing workload due to insufficient physical resources, leading to unavailability of services, especially when customers do not allow ample lead time for resource procurement.
Innovation Solution
Implementing a credit management system for resource sharing across multiple Cloud providers, allowing workload shifting and resource pooling through a fair-share algorithm that tracks and compensates resource usage with credits based on type, duration, and service level agreements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a Cloud provider uses only its own physical resources to fulfill service requests, then it maintains control over service quality and reliability, but it cannot meet resource requirements when customers do not allow ample lead time for resource procurement
Solution Approach 1:
The patent combines resources from multiple Cloud providers into a unified resource pool. When a service request exceeds the capacity of the primary Cloud provider, the system automatically merges with secondary providers to fulfill the shortfall, thereby maintaining service availability while gaining resource flexibility through partnerships.
Solution Approach 2:
The system establishes resource sharing agreements and pre-configures fallback mechanisms before resource shortages occur. By preliminarily setting up credit-based resource exchange frameworks and identifying alternative providers in advance, the system can rapidly respond to unexpected workload surges without requiring lead time for resource procurement.
2Productivity
If a Cloud provider procures additional physical resources to meet peak demand, then it can fulfill more service requests, but it increases infrastructure cost and resource idle time during low-demand periods
Solution Approach 1:
The system dynamically adjusts resource allocation based on real-time demand conditions. During peak periods, the primary Cloud provider utilizes its own resources and supplements with borrowed resources from partners. During low-demand periods, it returns excess resources to partners, thereby maintaining high service fulfillment capacity while minimizing resource idle time and associated costs.
Solution Approach 2:
The resource sharing framework enables Cloud providers to serve multiple functions: acting as primary providers during their peak periods and as backup providers for partners during their peak periods. This multi-functionality allows providers to maintain high productivity across the consortium while reducing individual infrastructure investment and idle resource time.
3Reliability
If a Cloud provider allows other providers to handle its workload, then it improves service availability during capacity shortages, but it introduces complexity in resource management and credit tracking
Solution Approach 1:
The patent introduces a centralized resource management system that acts as an intermediary between Cloud providers. This intermediary automatically handles workload routing, resource allocation, and credit tracking according to pre-established agreements. By delegating these complex management tasks to the intermediary system, individual providers can share workload while minimizing the complexity burden on each participant.
Solution Approach 2:
The system implements automated feedback mechanisms that continuously monitor resource usage, workload distribution, and credit balances. This real-time feedback enables dynamic adjustment of resource sharing decisions and automatic reconciliation of credits, thereby maintaining service availability while reducing the manual management complexity through automated control loops.
Data Source
AI summary
The present invention provides a solution for resource sharing (e.g., inter-Cloud) within a Cloud-computing environment. One objective of the present invention is to better utilize idle public or private Cloud infrastructures and improve the availability of Cloud services by allowing different Cloud service providers to virtually combine their services and infrastructures. In the event that there is not enough capacity for a single Cloud service provider to manage its workload, the workload may be shifted to additional infrastructures within the Cloud. The result of the workload shift may reduce the unavailability of Cloud services to the Cloud end-user by allowing another Cloud service provider to temporarily handle the workload. Based on the resource sharing activities, compensation (e.g., credits/tokens) can be exchanged between the participating Cloud providers to reflect their participation in the resource exchange.