Storage Tiering Bottleneck Resolution via Management Server

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Storage tiering and storage I/O control functionalities, when applied together, can lead to bottlenecks and performance issues due to misidentification of I/O levels and inappropriate data relocation between storage media, and de-duplication can cause unexpected performance decreases by altering access destinations from high to low performance storage.

Innovation Solution

A system and method that utilizes a management server to manage virtual volumes based on access frequency, allowing for the direct cancellation of access restrictions and reconfiguration of storage I/O control, and avoids de-duplication for pinned data to prevent performance degradation, ensuring optimal data placement and access.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If storage tiering functionality automatically moves data between storage media based on workload, then cost performance is improved, but bottlenecks occur when both storage tiering and storage I/O control functionalities are applied together

Engineering Contradiction:
Improvecost performanceVSAvoidperformance stability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent introduces a management server as an intermediary between the storage system and the host. The management server receives I/O control parameters from the host, translates them into storage tiering policies, and manages data relocation between storage media. This intermediary role resolves the conflict by coordinating both functionalities through a unified management layer, preventing bottlenecks while maintaining cost performance benefits.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent segments the storage management functionality into separate components: the storage system handles tiering operations while the host handles I/O control parameters. The management server acts as a bridge that translates between these segmented functions. This segmentation allows each component to operate independently without interfering with the other, resolving the bottleneck issue while preserving both functionalities.

Inventive Principle:
Principle #1Segmentation

2Quantity of substance

If de-duplication functionality detects multiple occurrences of the same data and replaces with pointer, then storage space is reduced, but unexpected performance decreases occur by altering access destinations

Engineering Contradiction:
Improvestorage spaceVSAvoidaccess speed
Core Design Contradiction:
Quantity of substanceVSSpeed

Solution Approach 1:

The patent implements dynamic de-duplication management where the system continuously monitors access patterns and dynamically adjusts de-duplication policies. When access speed degradation is detected, the system can dynamically disable de-duplication for specific data or revert to original access destinations. This dynamic adjustment resolves the contradiction by adapting the storage space optimization to maintain acceptable access performance.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent incorporates feedback mechanisms that monitor access performance and use this information to adjust de-duplication operations. When performance degradation is detected, the feedback loop triggers policy adjustments such as disabling de-duplication for affected data or restoring original access paths. This feedback-driven approach resolves the contradiction between storage space reduction and access speed maintenance.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS10050902B2Methods and apparatus for de-duplication and host based QoS in tiered storage system
Publication Date: 2018.08.14 HITACHI VANTARA LTD
  • US10050902B2 patent drawing
  • US10050902B2 patent drawing
  • US10050902B2 patent drawing

AI summary

Example implementations described herein are directed to addressing potential decreases in performance for a storage system coupled to a server. In a first example implementation, systems and methods are directed to addressing potential performance issues for storage tiering and the storage I/O control. In a second example implementation, systems and methods are directed to addressing potential performance issues for storage tiering and de-duplication.