Virtual Object Force Provisioning in Degraded Stretched Clusters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing approaches to force provisioning virtual objects in degraded stretched clusters either fail to provision due to infrastructure failures or provide inadequate protection, leading to inefficiencies and increased operational costs during transient workloads.
Innovation Solution
Implementing a 'smart forced provisioning' method that ensures virtual objects are provisioned with at least one full RAID tree in a healthy site of a degraded stretched cluster, even if the cluster is partially failed, using RAID-0 in the failed site and RAID-1 in the healthy site, thereby maintaining high protection levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If force provisioning is implemented in degraded stretched clusters using previous approaches, then virtual objects can be provisioned despite infrastructure failures, but the objects receive inadequate protection with low levels of redundancy
Solution Approach 1:
The patent applies local quality by differentiating the provisioning approach between healthy sites and failed sites. In healthy sites, full RAID trees with appropriate redundancy (RAID-1 or RAID-5/6) are provisioned to ensure high protection levels. In failed sites, RAID-0 is used temporarily to enable provisioning while accepting lower protection. This localized differentiation allows the system to maintain high reliability where possible while ensuring productivity where failures occurred.
Solution Approach 2:
The system dynamically adjusts the RAID configuration based on the health status of cluster sites. When provisioning virtual objects in degraded stretched clusters, the system automatically detects which sites are healthy and which have failures, then applies appropriate RAID configurations accordingly. This dynamic adaptation allows the system to optimize between protection and provisioning success based on real-time cluster conditions.
2Reliability
If virtual objects are provisioned with full RAID trees in all sites, then high protection levels are maintained, but provisioning fails when storage policy resources are insufficient in degraded sites
Solution Approach 1:
The patent implements partial action by provisioning full RAID trees with high protection only in healthy sites where sufficient storage resources are available. In failed sites where resources are insufficient, the system provisiones only RAID-0 configurations temporarily. This partial approach ensures that the portions of the system where full protection is feasible receive it, while accepting reduced protection in constrained areas to achieve overall provisioning success.
3Reliability
If RAID-1 configuration is used in degraded sites with insufficient resources, then protection is maintained, but provisioning cannot be completed due to resource constraints
Solution Approach 1:
The system applies partial action by implementing RAID-1 redundancy only in healthy sites where resources are sufficient. In degraded sites with insufficient resources, the system accepts partial provisioning with RAID-0 (no redundancy) temporarily. This approach ensures that redundancy is applied where feasible rather than attempting to apply it universally, which would fail due to resource constraints.
Solution Approach 2:
The provisioning system dynamically determines the appropriate RAID configuration based on available resources in each site. When resources are insufficient for RAID-1, the system automatically adjusts to RAID-0 in affected sites while maintaining RAID-1 in healthy sites. This dynamic resource-based configuration ensures provisioning feasibility while maximizing protection where possible.
Data Source
AI summary
The present disclosure is related to methods, systems, and machine-readable media for force provisioning virtual objects in degraded stretched clusters. A request to provision a virtual object by a stretched cluster according to a storage policy specified as part of the request can be received by a software defined data center (SDDC). The cluster can include a plurality of sites. An insufficiency of storage policy resources to satisfy the storage policy specified for the virtual object can be determined. The virtual object can be force provisioned responsive to determining storage policy resources sufficient to satisfy the storage policy at one of the plurality of sites.