Workload-Aware Storage Allocation via Predictive Pool Migration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current thin-provisioning storage-management systems perform dynamic storage-allocation and load-balancing functions independently, leading to less accurate and less efficient storage allocation, as they do not fully account for the intrinsic relationships between workload-balancing tasks and capacity-allocation tasks.
Innovation Solution
A workload-aware thin-provisioning mechanism that integrates storage allocation and load-balancing functions by predicting future storage and workload needs based on historical profiles and physical capacity constraints, allowing for the dynamic reallocation of physical storage volumes between pools to prevent conflicts and optimize resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If dynamic storage-allocation and load-balancing functions are performed independently by two different components, then the system structure is simpler and easier to implement, but the storage allocation accuracy and efficiency deteriorates
Solution Approach 1:
The patent merges the storage-allocation component and load-balancing component into a single integrated component that performs both functions simultaneously. This integration allows the system to consider workload information when making storage-allocation decisions, thereby improving storage allocation accuracy while maintaining manageable system complexity through a unified approach.
2Reliability
If thin-provisioning allocates enough real capacity to ensure users do not run out of physical storage, then storage availability is improved, but physical storage capacity waste increases
Solution Approach 1:
The patent implements a feedback mechanism where the integrated component continuously monitors actual storage consumption patterns and workload characteristics. Based on this feedback, the system dynamically adjusts thin-provisioning allocation decisions, allocating physical storage capacity more precisely according to actual needs rather than using fixed conservative estimates, thereby reducing waste while maintaining availability.
Solution Approach 2:
The system transitions from static storage allocation based on fixed estimates to dynamic allocation that adapts to changing workload patterns and actual consumption rates. The integrated component can reallocate physical storage capacity in real-time based on current system state, optimizing the balance between availability and waste reduction.
3Loss of substance
If thin-provisioning uses conservative allocation estimates, then physical storage capacity waste is reduced, but the frequency of reprovisioning tasks increases
Solution Approach 1:
The feedback mechanism monitors actual storage consumption in real-time and triggers reprovisioning tasks only when necessary, based on actual usage patterns rather than conservative periodic re-allocation. This reduces the frequency of unnecessary reprovisioning tasks while maintaining optimal storage capacity utilization.
Solution Approach 2:
The system performs preliminary analysis of workload patterns and consumption rates to predict future storage needs, allowing proactive optimization of allocation decisions before reprovisioning becomes necessary. This reduces the frequency of reactive reprovisioning tasks by maintaining more accurate forward-looking allocation estimates.
4Measurement precision
If storage-allocation and load-balancing are integrated into a single operation, then storage allocation accuracy and efficiency is improved, but the device complexity increases
Solution Approach 1:
The patent combines storage-allocation and load-balancing functions in a single integrated component that processes both storage capacity and workload information simultaneously. This merger improves storage allocation accuracy by considering workload characteristics, while the unified structure avoids the overhead of coordinating multiple separate components, keeping system complexity manageable.
Data Source
AI summary
A method and associated systems for a workload-aware thin-provisioning system that allocates physical storage to virtual resources from pools of physical storage volumes. The system receives constraints that limit the amount of storage that can be allocated from each pool and the total workload that can be directed to each pool. It also receives lists of previous workloads and allocations associated with each volume at specific times in the past. The system then predicts future workloads and allocation requirements for each volume by regressing linear equations derived from the received data. If the predicted values indicate that a pool will at a future time violate a received constraint, the system computes the minimum costs to move each volume of the offending pool to a less-burdened pool. It then selects the lowest-cost combination of volume and destination pool and then moves the selected volume to the selected pool.