Priority-Based Rebuild Orchestration for HCI Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In hyper-converged infrastructure (HCI) deployments, the rebuild process of storage object components is inefficient due to random scheduling and limited parallel rebuilding, which can lead to critical VMs being delayed and vulnerable to further failures, potentially causing service outages or data loss.
Innovation Solution
Implementing a priority-based rebuild orchestration system that assigns priority classes to virtual machines (VMs) based on their storage policies, ensuring that higher priority VMs are rebuilt first, thereby reducing the time to compliance with fault tolerance requirements and minimizing data loss risks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If random scheduling is used for rebuilding storage object components, then all components can be rebuilt eventually, but critical VMs may be delayed and remain vulnerable to further failures
Solution Approach 1:
The patent changes the scheduling parameter from random to priority-based, where each VM is assigned a priority level (e.g., P0-highest, P1-high, P2-medium, P3-lowest). This parameter change ensures that critical VMs are rebuilt first, reducing their vulnerability window while maintaining systematic rebuild coverage for all components
Solution Approach 2:
The patent performs preliminary classification of VMs into priority groups before the rebuild process begins. By pre-assigning priority levels based on business criticality and storage policies, the system prepares the rebuild schedule in advance, ensuring that high-priority VMs are addressed first without delaying low-priority ones
2Use of energy by moving object
If limited parallel rebuilding is implemented, then system resources are conserved, but the overall rebuild speed and compliance time are reduced
Solution Approach 1:
The patent implements dynamic parallel rebuilding where the number of concurrent rebuild operations is adjusted based on system conditions. Multiple components can be rebuilt in parallel up to a threshold, and the system dynamically manages the balance between resource consumption and rebuild speed by allowing flexible parallelization within resource constraints
3Device complexity
If no priority classification is applied, then the rebuild process is simple to implement, but business-critical VMs remain vulnerable to failures for extended periods
Solution Approach 1:
The patent segments the VM population into distinct priority groups (P0-P3) based on business criticality and storage policies. This segmentation allows the rebuild system to process different groups in a structured manner, improving service continuity for critical VMs while maintaining manageable complexity through clear categorization rules
Data Source
AI summary
Techniques for orchestrating and prioritizing the rebuild of storage object components in a hyper-converged infrastructure (HCI) deployment comprising a cluster of host systems are provided. In one set of embodiments, a computer system can identify a list of storage object components impacted by a maintenance event or failure of a host system in the cluster. The computer system can further determine a priority class for each storage object component in the list, where the determined priority class is based on a virtual machine (VM)-level priority class assigned to a VM to which the storage object component belongs. The computer system can then initiate rebuilds of the storage object components in the list on a per-VM and per-priority class basis, such that: (1) the rebuilds of storage object components belonging to the same VM are initiated consecutively, and (2) the rebuilds of storage object components with higher priority classes are initiated before the rebuilds of storage object components with lower priority classes.