Storage Tier Selection for Recovery Time Objectives
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing DRaaS solutions face challenges in predicting Recovery Time Objective (RTO) requests, leading to overprovisioning of resources by Service Providers and concerns about reliability for Service Consumers, which discourages them from using DRaaS solutions.
Innovation Solution
The implementation requires Service Consumers to explicitly specify their RTO, compares this with the restore times of available storage tiers, and stores the alternate volume in the tier with a restore time less than the RTO, while prioritizing the cheapest option if multiple tiers satisfy the RTO.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If Service Providers overprovision resources to meet RTO requests, then RTO satisfaction reliability is improved, but resource cost increases
Solution Approach 1:
The system dynamically selects storage tiers based on the specific RTO requirement of each service consumer. Instead of statically overprovisioning all resources to the highest tier, the system adapts the storage tier selection to match each consumer's RTO needs, optimizing the balance between reliability and resource cost.
Solution Approach 2:
The system changes the parameter of storage tier selection based on the RTO parameter specified by service consumers. By varying the storage tier (from high-performance to cost-effective tiers) according to the RTO requirement, the system achieves reliable RTO satisfaction without uniform overprovisioning of resources.
2Loss of time
If Service Providers use high-performing storage tiers to minimize RTO, then RTO is reduced, but service cost increases
Solution Approach 1:
The system applies local quality by matching different storage tier qualities to different RTO requirements. Service consumers with stringent RTO needs receive high-performing storage, while those with more flexible RTO requirements are served by cost-effective storage tiers, optimizing the quality-cost balance locally for each consumer.
Solution Approach 2:
The system changes the storage performance parameter based on the RTO parameter. By adjusting the storage tier performance level to match the required RTO, the system achieves the necessary time loss reduction without incurring the cost of uniformly high-performance storage across all services.
3Reliability
If Service Consumers prepare their own infrastructure to ensure RTO reliability, then RTO reliability is improved, but Service Provider revenue decreases
Solution Approach 1:
The system enables self-service by allowing service consumers to specify their own RTO requirements. This empowers consumers to define their reliability needs while the Service Provider automatically configures appropriate storage tiers, eliminating the need for consumers to build their own infrastructure while maintaining RTO reliability.
Solution Approach 2:
The system uses feedback from service consumers regarding their RTO requirements to automatically select appropriate storage tiers. This feedback loop ensures that RTO reliability is maintained according to consumer needs while the Service Provider optimizes resource allocation and revenue based on actual requirements rather than worst-case assumptions.
Data Source
AI summary
Systems and methods described herein can involve, for a request to create a recovery volume having a specified recovery time objective, identifying a tier from a storage system managing a plurality of tiers that has a restore time that satisfies the specified recovery time objective; and creating the recovery volume in the identified tier of the storage system in response to the request.