Probabilistic Offload Engine for Hierarchical Object Storage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current storage solutions, such as SAN and NAS, are inefficient in managing storage resources due to their block-based architecture, which hinders storage optimization based on file or object concepts, leading to issues like wasted space and maintenance downtime, and fail to effectively manage large-scale storage needs.

Innovation Solution

A probabilistic offload engine for distributed hierarchical object storage devices that uses key/value-based storage systems to predict access patterns and move data between high-cost and low-cost storage layers, leveraging metadata for intelligent storage management and synchronization between storage tiers.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If block-based storage architecture (SAN/NAS) is used, then centralized storage management is achieved, but storage optimization based on file/object concepts cannot be performed, leading to wasted space and maintenance downtime

Engineering Contradiction:
Improvecentralized storage managementVSAvoidstorage optimization capability
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent segments storage into hierarchical tiers (hot, warm, cold storage) with different access characteristics and cost profiles. This segmentation enables object-based optimization while maintaining centralized management, resolving the contradiction between ease of operation and storage optimization capability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically migrates objects between storage tiers based on access patterns and cost considerations. This dynamic behavior allows the system to optimize storage usage in real-time while maintaining centralized control, addressing both ease of operation and productivity requirements.

Inventive Principle:
Principle #15Dynamics

2Quantity of substance

If thin provisioning is implemented, then storage space over-allocation is achieved, but file systems cannot reuse past blocks and waste space that cannot be reclaimed online

Engineering Contradiction:
Improvestorage space utilizationVSAvoidspace reclamation efficiency
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent implements feedback mechanisms that monitor object access patterns and storage usage in real-time. This feedback enables the system to dynamically adjust storage allocation and reclamation strategies, improving both space utilization and reclamation efficiency without compromising file system performance.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system changes storage parameters dynamically based on access patterns, transitioning objects between different storage tiers. This parameter change approach allows for efficient space utilization while maintaining the ability to reclaim space online, resolving the contradiction between quantity of substance and productivity.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If data is moved between high-cost and low-cost storage layers, then storage performance and capacity are optimized, but data transfer time and complexity increase

Engineering Contradiction:
Improvestorage performanceVSAvoiddata transfer time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent performs preliminary actions by pre-classifying objects into storage tiers based on their access patterns and characteristics. This pre-processing reduces the complexity and time required for subsequent data migration operations, optimizing both performance and transfer time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements self-service mechanisms where objects automatically migrate between storage tiers based on predefined policies and access patterns. This autonomous behavior minimizes manual intervention and reduces transfer complexity while maintaining optimized performance.

Inventive Principle:
Principle #25Self-service

4Adaptability or versatility

If probabilistic algorithms are used to predict access patterns, then storage management intelligence is improved, but algorithm complexity and computational overhead increase

Engineering Contradiction:
Improveaccess pattern prediction accuracyVSAvoidalgorithm complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent changes algorithm parameters dynamically based on system conditions and data characteristics. This adaptive parameter adjustment improves prediction accuracy while controlling computational overhead, resolving the contradiction between adaptability and device complexity.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system applies partial action by using probabilistic algorithms only for critical decision-making points rather than continuously. This selective application reduces computational overhead while maintaining sufficient prediction accuracy for effective storage management.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP2510455B1Probabilistic offload engine for distributed hierarchical object storage devices
Publication Date: 2019.07.31 SCALITY
  • EP2510455B1 patent drawingFigure 1
  • EP2510455B1 patent drawingFigure 2
  • EP2510455B1 patent drawingFigure 3

AI summary

A method and system having a probabilistic offload engine for distributed hierarchical object storage devices is disclosed. According to one embodiment, a system comprises a first storage system and a second storage system in communication with the first storage system. The first storage system and the second storage system are key/value based object storage devices that store and serve objects. The first storage system and the second storage system execute a probabilistic algorithm to predict access patterns. The first storage system and the second storage system execute a probabilistic algorithm to predict access patterns and minimize data transfers between the first storage system and the second storage system.