Thin Clone Shared Image Storage for High Availability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The high cost and inefficiency of implementing clustered file systems for managing shared images and write spaces in on-demand desktop streaming (ODDS) systems make high-availability solutions economically infeasible.

Innovation Solution

A method and system that involve creating a thin clone of a shared image on a storage array, using a partition, where a provisioning server manages input/output operations, allowing clients to access a shared image with user-specific write spaces without the need for a clustered file system, thereby reducing costs and improving availability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a clustered file system is used to manage shared images and write spaces for high availability, then reliability is improved, but device complexity and cost increase significantly

Engineering Contradiction:
Improvehigh availabilityVSAvoidclustered file system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

A storage array with logical units acts as an intermediary between multiple provisioning servers and clients. The storage array implements redundancy and failover mechanisms internally, allowing provisioning servers to access shared images without requiring complex clustered file systems. This mediator handles the complexity of high-availability management, simplifying the overall system architecture.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system creates thin clones of shared images stored on the storage array for each provisioning server. These clones allow multiple servers to access identical image data without requiring synchronized file systems. The thin clone technology reduces storage overhead while enabling high availability through automatic failover to alternative clones if needed.

Inventive Principle:
Principle #26Copying

2Reliability

If multiple provisioning servers are deployed for operational redundancy, then reliability is improved, but the cost and complexity of managing clustered file systems increase

Engineering Contradiction:
Improveoperational redundancyVSAvoidclustered file system management
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

Each provisioning server maintains a thin clone of the shared image on the storage array. These clones are lightweight copies that enable independent operation of each provisioning server. If one server fails, another server can immediately take over using its own clone, providing operational redundancy without requiring complex clustered file system coordination.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The storage array serves multiple functions simultaneously: it stores shared images, creates thin clones for each provisioning server, manages write spaces, and provides failover capabilities. This multi-functional approach eliminates the need for separate clustered file system infrastructure, reducing overall system complexity while maintaining reliability.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Quantity of substance

If thin clones are used instead of full clones for write spaces, then storage efficiency is improved, but data management complexity increases

Engineering Contradiction:
Improvestorage spaceVSAvoidthin clone management
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The storage array automatically manages thin clone creation, maintenance, and updates without requiring complex external management systems. When data is written to a provisioning server's write space, the storage array handles the cloning and data distribution transparently. This self-managing approach reduces the apparent complexity for provisioning servers while maximizing storage efficiency through thin provisioning.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS8069217B2System and method for providing access to a shared system image
Publication Date: 2011.11.29 DELL PROD LP
  • US8069217B2 patent drawing
  • US8069217B2 patent drawing
  • US8069217B2 patent drawing

AI summary

Systems and methods for providing data integrity for stored data are disclosed. A method for providing access to a remote shared image may include receiving from a first provisioning server a request to create a write space for a client communicatively coupled to the first provisioning server. The message may further include creating a partition on a first logical unit disposed in a storage array, the first logical unit communicatively coupled to the first provisioning server and associated with the client. The method may additionally include storing a thin clone of a shared image on the partition, the shared image common to the client and at least one other client and stored on a second logical unit, the thin clone comprising a write space for data specific to the client.