Virtual Machine Storage Reclamation Without Shutdown

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional storage reclamation in virtual environments is inefficient as it typically requires shutting down virtual machines to reclaim physical storage resources, leading to performance impacts and resource wastage, especially when virtual machines are left running for extended periods.

Innovation Solution

A method for performing storage reclamation of physical resources while the virtual machine is running, using a reclamation communication protocol that translates virtual storage reclamation indications into physical storage resource locations for efficient resource recovery, allowing for continuous operation with minimal overhead.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of substance

If conventional storage reclamation is performed by shutting down virtual machines, then physical storage resources can be reclaimed, but virtual machine performance and operational continuity are adversely impacted

Engineering Contradiction:
Improvephysical storage resource reclamationVSAvoidvirtual machine operational continuity
Core Design Contradiction:
Loss of substanceVSReliability

Solution Approach 1:

The system performs preliminary identification of reclaimable storage blocks by analyzing virtual machine deletion operations before actual reclamation occurs. The hypervisor tracks deleted virtual storage blocks and prepares reclamation candidates, allowing the virtual machine to continue running while the reclamation process is staged and ready for execution with minimal disruption.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The hypervisor acts as an intermediary layer between the virtual machine and physical storage resources. It intercepts storage operations, identifies deletion opportunities, and manages the reclamation process independently, allowing the virtual machine to operate uninterrupted while physical storage is reclaimed through the hypervisor's mediation without requiring VM shutdown.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If virtual machines are left running for extended periods to maintain performance, then operational results are achieved, but significant storage resources are consumed and cannot be readily reclaimed

Engineering Contradiction:
Improvevirtual machine operational performanceVSAvoidstorage resource consumption
Core Design Contradiction:
ProductivityVSLoss of substance

Solution Approach 1:

The system implements feedback mechanisms where the hypervisor continuously monitors virtual machine storage operations, tracks deleted blocks, and provides real-time information about reclaimable resources. This feedback loop enables dynamic storage management that adapts to virtual machine activity patterns, allowing reclamation to occur based on actual usage patterns without impacting performance.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The virtual machine effectively serves itself by generating the reclamation opportunities through its own deletion operations. The hypervisor leverages the virtual machine's natural write and delete patterns to identify reclamation candidates, turning the virtual machine's operational activity into the mechanism that enables its own storage optimization without external intervention or shutdown.

Inventive Principle:
Principle #25Self-service

3Loss of substance

If traditional storage reclamation approaches are used, then physical storage can be recovered in some situations, but the process requires virtual machine shutdown and complete image deletion which is inefficient

Engineering Contradiction:
Improvephysical storage recoveryVSAvoidreclamation process time
Core Design Contradiction:
Loss of substanceVSLoss of time

Solution Approach 1:

The reclamation process is segmented into independent stages: identification of deleted virtual blocks, translation to physical block addresses, and actual reclamation execution. This segmentation allows each stage to be performed independently and asynchronously, enabling the virtual machine to continue running while the reclamation progresses through different stages without requiring complete shutdown or atomic operations.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transitions from static, all-or-nothing reclamation approaches to dynamic, incremental reclamation. The hypervisor continuously identifies and reclaims storage blocks in small units as deletion opportunities arise, allowing the reclamation process to adapt dynamically to virtual machine activity patterns and proceed continuously without interruption rather than requiring periodic shutdowns.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS8813071B2Storage reclamation systems and methods
Publication Date: 2014.08.19 ARCTERA US LLC
  • US8813071B2 patent drawing
  • US8813071B2 patent drawing
  • US8813071B2 patent drawing

AI summary

Efficient and effective storage reclamation systems and methods are presented. In one embodiment, a storage reclamation method comprises: receiving an indication from a virtual machine that a reclamation opportunity for reclamation of a physical storage resource is available; and performing a reclamation process of the physical storage resource in accordance with the indication from the virtual machine that the reclamation opportunity is available, wherein the reclamation process is performed while the virtual machine is running. The indication can be received from a virtual communication protocol (e.g., can include an in-band communication protocol, out-of band communication protocol, a socket based communication protocol or a serial character device communication protocol).