Disk Slice Distribution for Virtual Storage Over-Commitment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In virtualized computing environments, over-commitment of storage space can lead to performance degradation as heavily utilized physical disks exhaust their capacity, causing imbalance and impacting the overall performance of virtual disks.
Innovation Solution
A storage manager monitors and predicts allocation rates for virtual disks, evaluates physical disk utilization, and balances storage by splitting data into multiple disk slices and redistributing them across available physical disks to prevent over-commitment and optimize performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If storage space is over-committed to virtual disks, then storage resource utilization is improved, but system reliability deteriorates when physical disks exhaust capacity
Solution Approach 1:
The patent divides virtual disk storage into multiple slices that can be independently distributed across different physical disks. This segmentation allows the storage system to balance data across multiple physical disks, preventing any single disk from becoming a bottleneck or failure point, thereby maintaining reliability while improving overall utilization.
Solution Approach 2:
The patent implements dynamic distribution of disk slices that can be automatically rebalanced based on current physical disk utilization. When physical disks become unbalanced or reach capacity thresholds, the system dynamically migrates slices between disks, ensuring continuous reliability adaptation while maximizing storage resource utilization.
2Device complexity
If data is concentrated on fewer physical disks, then device complexity is reduced, but performance deteriorates due to uneven utilization
Solution Approach 1:
The patent implements self-service automated balancing that monitors physical disk utilization and automatically redistributes disk slices without manual intervention. This maintains low operational complexity while ensuring optimal performance through continuous automatic rebalancing of storage resources.
Solution Approach 2:
The patent changes the distribution parameters of disk slices across physical disks based on utilization metrics. By dynamically adjusting distribution parameters rather than maintaining fixed allocations, the system achieves balanced performance across all disks without requiring complex manual management.
3Stability of the object's composition
If disk slices are redistributed across multiple physical disks, then storage balance is improved, but operation time increases due to data migration
Solution Approach 1:
The patent applies partial action by performing incremental rebalancing operations that migrate only the necessary portion of data required to achieve balance, rather than complete data migrations. This approach achieves storage balance while minimizing the time loss associated with data migration operations.
Solution Approach 2:
The patent implements periodic monitoring and rebalancing operations that adjust disk slice distribution at scheduled intervals or based on utilization thresholds. This periodic approach maintains storage balance over time while avoiding continuous data migration, thereby reducing overall operation time.
Data Source
AI summary
An example method for balancing disk usages in a virtualized computing environment may include generating, by a storage manager configured to manage the virtualized computing environment, predicted-disk-utilizations for a plurality of physical disks provisioned to a virtual disk in the virtualized computing environment. In response to a determination that a first predicted-disk-utilization for a first physical disk selected from the plurality of physical disks exceeds a predetermined-utilization-threshold, the storage manager may generate a storage-balancing policy for splitting the virtual disk into a number of disk slices and distributing the number of disk slices to a subset of candidate physical disks selected from the plurality of physical disks, wherein the storage-balancing policy provides a minimum data-movement-variance for distributing of the number of disk slices to the subset of the candidate physical disks.