Virtual Image Storage Allocation Across Block, File, and Object Tiers

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Information handling systems employing virtualization face challenges in scalability, performance, and latency due to demanding storage requirements for thousands of user sessions and bursty workloads, which hamper storage backend performance and capacity.

Innovation Solution

A storage management system that allocates elements of a virtual image across block-based, file-based, and object-based storage media based on characterization, enabling tiering and scaling for optimized performance and capacity, and providing unified management with a common platform.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a storage back-end hosts thousands of user sessions in a virtualized environment, then scalability is improved, but storage backend performance and capacity are hampered

Engineering Contradiction:

ImprovescalabilityVSAvoidstorage backend performance

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The storage system is segmented into multiple storage tiers (block storage, file storage, object storage) with different performance characteristics and capacity profiles. This allows the system to handle thousands of user sessions by distributing workloads across tiers rather than relying on a single uniform storage backend, thereby maintaining scalability while preserving performance.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Different storage tiers are assigned to different workload characteristics based on local quality principles. High-performance block storage is allocated for latency-sensitive applications, while high-capacity object storage handles bulk data archiving. This localized optimization ensures that each storage tier operates within its optimal performance envelope, resolving the contradiction between scalability and performance.

Inventive Principle:

Principle #3Local quality

2Adaptability or versatility

If storage back-end supports bursty client workloads, then adaptability is improved, but latency in application responsiveness increases

Engineering Contradiction:

Improveworkload adaptabilityVSAvoidapplication responsiveness latency

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The storage system implements dynamic workload distribution that adapts to bursty client requests in real-time. When bursty workloads are detected, the system dynamically shifts data access patterns between storage tiers, utilizing high-speed block storage for immediate response and asynchronously caching data in object storage, thereby maintaining adaptability while minimizing latency.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system performs preliminary actions by pre-loading frequently accessed data into high-performance block storage caches and maintaining warm storage tiers ready for immediate access. This preparation before actual workload bursts reduces the latency impact when bursts occur, allowing the system to maintain both adaptability and responsiveness.

Inventive Principle:

Principle #10Preliminary action

3Ease of operation

If virtualization migrates client workloads to data center, then manageability and security are improved, but storage requirements and complexity increase

Engineering Contradiction:

ImprovemanageabilityVSAvoidstorage management complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The storage management system provides universal multi-functional capabilities through a unified interface that manages block, file, and object storage tiers simultaneously. This universal manager consolidates what would otherwise be separate complex management systems into a single coherent platform, improving manageability while controlling complexity through standardization.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The storage system implements self-service automation where the management system automatically characterizes workloads, selects appropriate storage tiers, and performs data migration without manual intervention. This self-service approach reduces the operational complexity burden on administrators while maintaining the manageability benefits of centralized virtualized storage.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS8793308B2Systems and methods for managing storage for client workloads

Publication Date: 2014.07.29 DELL PROD LP

AI summary

A method for managing the storage of a client workload is disclosed. A storage management system is provided. The storage management system is operable to communicatively connect to a remote client computing device via a network. The storage management system includes a block-based storage medium, a file-based storage medium, and an object-based storage medium. A virtual image, which is associated with a remote client computing device, is stored. Elements of the virtual image are allocated across one or more of the block-based storage medium, the file-based storage medium, and the object-based storage medium, where the allocation depends on a characterization of the elements.