Dynamic Redundancy Determination for Cloud Bursting Cost Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud bursting technologies often incur unnecessary costs by deploying systems to multiple public cloud sites regardless of necessity, leading to inefficient resource allocation and increased expenses.
Innovation Solution
A redundancy determination system that deploys systems to multiple public cloud sites only when the recovery time objective (RTO) is less than the build-out time, and to a single site when RTO exceeds the build-out time, optimizing resource allocation and reducing costs by determining the necessary deployment based on operational status and cost-effectiveness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system is deployed to the plurality of sites regardless of whether or not it is necessary, then the system can continue processing even in case of failure, but an unnecessary cost increases
Solution Approach 1:
The patent changes the parameter of deployment configuration from fixed multi-site to dynamic selection between single-site and multi-site based on the RTO parameter. By making the deployment parameter conditional rather than fixed, the system achieves cost optimization while maintaining reliability when necessary.
Solution Approach 2:
The patent introduces dynamic decision-making based on comparing RTO with build-out time. The deployment strategy is no longer static but adapts based on system requirements, transitioning between single-site and multi-site deployment modes to optimize the balance between reliability and cost.
2Reliability
If the system is deployed to the plurality of sites, then the processing can be continued even in a case where a failure occurs, but an unnecessary cost may increase
Solution Approach 1:
The patent changes the resource allocation parameter from always deploying to multiple sites to conditionally deploying based on RTO comparison. This parameter change enables the system to allocate resources efficiently - using multi-site deployment only when fault tolerance is critical and single-site deployment when it is not necessary.
Solution Approach 2:
The patent applies partial action by deploying to multiple sites only when necessary (when RTO < build-out time) rather than always deploying to multiple sites. This partial deployment approach avoids excessive resource allocation while maintaining adequate fault tolerance for systems that require it.
3Reliability
If cloud bursting is performed by deploying to multiple sites, then system continuity is ensured, but cost efficiency decreases due to unnecessary deployment
Solution Approach 1:
The patent changes the cloud bursting deployment parameter from fixed multi-site to dynamic selection based on RTO. This parameter change resolves the contradiction by enabling cost-efficient single-site deployment for systems where continuity can be achieved through other means, while reserving multi-site deployment for systems where it is truly necessary for continuity.
Solution Approach 2:
The patent introduces dynamic deployment strategy selection that adapts to different system requirements. By making the deployment configuration dynamic rather than static, the system can optimize for cost efficiency in appropriate cases while maintaining system continuity where required, thus resolving the productivity-cost efficiency contradiction.
Data Source
AI summary
Provided is a redundancy determination system extending a system on an on-premise data center to a public cloud by using a computer including a processor and a memory, in which the processor performs first processing of deploying the system in a plurality of of sites included in the public cloud when a recovery time objective of the system is less than a build-out time of the system that is calculated on the basis of an operational status of the system, and second processing of deploying the system in one site of the sites included in the public cloud when the recovery time objective of the system is not less than the build-out time.