Virtual Storage Volume Combining Performance and Capacity Tiers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage tiering systems do not efficiently utilize higher performing tiers, leading to increased costs and limited storage performance due to inefficient data movement between storage tiers based on access frequency.
Innovation Solution
The creation of a virtual storage volume that combines performance volumes and capacity volumes, allowing input-output requests to be processed using the performance tier while utilizing the capacity tier for storage, thereby optimizing data placement and movement based on usage patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is moved between storage tiers based on access frequency, then storage capacity is optimized, but storage performance deteriorates due to inefficient data movement
Solution Approach 1:
The storage volume is segmented into a performance portion and a capacity portion, allowing different data to be stored in different tiers simultaneously. This segmentation enables the system to maintain high-performance access for frequently accessed data while efficiently utilizing capacity tier for less accessed data, resolving the contradiction between capacity optimization and performance maintenance.
Solution Approach 2:
A storage controller acts as an intermediary between the host and the tiered storage system, managing data placement and movement between tiers. The controller intelligently determines which data resides in which tier based on access patterns, providing transparent high-performance access while optimizing capacity utilization without requiring data movement to impact performance.
2Productivity
If higher performing storage tiers are utilized, then storage performance improves, but costs increase
Solution Approach 1:
Different portions of the storage volume have different quality characteristics - the performance portion provides high-speed access while the capacity portion provides cost-effective storage. This local quality differentiation allows the system to deliver high performance where needed while minimizing costs for bulk storage, resolving the contradiction between performance and cost.
Solution Approach 2:
The system dynamically changes the parameters of data placement by modifying which data resides in which tier based on access patterns. This parameter change enables the system to adapt to varying workloads, maintaining high performance for active data while cost-effectively storing inactive data in lower-tier storage.
3Productivity
If data is moved between storage tiers, then storage efficiency improves, but access time increases due to data movement operations
Solution Approach 1:
The storage controller performs preliminary actions by proactively managing data placement in the performance and capacity portions based on predicted access patterns. This preliminary organization of data ensures that frequently accessed data is already in the performance tier before access is needed, eliminating data movement delays during actual access operations and resolving the contradiction between efficiency and access time.
Data Source
AI summary
Methods, apparatus, and processor-readable storage media are provided herein for storage volume creation using performance volumes and capacity volumes. An example computer-implemented method includes configuring a storage system with at least first and second storage tiers each comprising a plurality of storage devices; creating a virtual storage volume having a first portion corresponding to at least a portion of the plurality of storage devices of the first storage tier and a second portion corresponding to at least a portion of the plurality of storage devices of the second storage tier; and processing input-output requests from one or more host devices associated with the virtual storage volume, wherein the processing comprises moving data between the first portion and the second portion of the virtual storage volume so that each of the input-output requests is processed using the first portion of the virtual storage volume.


