SAN Storage Reclamation for Virtual Containers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current storage area networks (SANs) lack a simple mechanism for identifying and reclaiming unused virtual blocks, leading to underutilized storage capacity and peak limits in virtual data containers.
Innovation Solution
A SAN reclamation system that monitors containers for unused volumes exceeding an empty capacity threshold and idle time thresholds, automatically separating and un-mapping these volumes from the physical host to return them to the storage pool, thereby optimizing storage resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If storage capacity is allocated to virtual data containers, then storage virtualization is achieved, but unused virtual blocks cannot be identified and reclaimed
Solution Approach 1:
The system enables self-service reclamation by implementing automated monitoring and identification of unused virtual blocks within virtual data containers. The reclamation system autonomously detects unused blocks, determines their reclamation status, and returns them to the storage pool without requiring manual intervention, thus resolving the complexity of implementing a reclamation mechanism while maintaining storage virtualization.
2Productivity
If manual identification of unused blocks is attempted, then reclamation can occur, but the process becomes complex and inefficient
Solution Approach 1:
The system implements feedback mechanisms by continuously monitoring virtual data containers for unused blocks and automatically initiating reclamation processes when unused blocks are detected. This automated feedback loop eliminates the need for complex manual identification processes while improving reclamation efficiency, as the system self-regulates based on real-time storage container status.
Data Source
AI summary
A system, method and program product for reclaiming storage volumes in a storage area network (SAN). A method is disclosed that includes: monitoring each of a set containers implemented by a physical host, wherein each container holds a set of storage volumes that are mapped to a storage pool via a SAN switch; determining whether unused volumes in a selected container exceed an empty capacity threshold; in response to the empty capacity threshold being exceeded, determining whether an idle time threshold for the unused volumes has been exceeded with no I/O activity; and in response to the idle time threshold being exceeded with no I/O activity, separating the unused volumes from the selected container, un-mapping the unused volumes from the physical host, and returning the unused volumes to the storage pool.