Distributed Storage Management Hierarchy for Latency Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In enterprise storage systems, disruptions occur when information about the storage system is gathered but not promptly acted upon, leading to inefficiencies and potential issues in managing storage array controllers and devices.

Innovation Solution

Implementing a distributed management system where each storage system element has a distributed manager to gather information, determine actions, and execute them, along with a hierarchical organization of these managers to ensure efficient and coordinated management across multiple storage devices and controllers.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If a primary storage array controller gathers information about the storage system and initiates actions in dependence upon the gathered information, then centralized control and coordination are achieved, but latency increases and management efficiency deteriorates due to the time delay between information gathering and action initiation

Engineering Contradiction:
Improvelatency between information gathering and action initiationVSAvoidmanagement efficiency
Core Design Contradiction:
Loss of timeVSProductivity

Solution Approach 1:

The patent segments the centralized control function into distributed control mechanisms across multiple storage array controllers. Each controller can independently gather information and initiate actions within its jurisdiction, eliminating the sequential delay inherent in centralized processing. This segmentation allows parallel information gathering and action initiation across multiple controllers simultaneously.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent merges the information gathering and action initiation functions into a single coordinated process among multiple controllers. Rather than separating these functions across different controllers (which would increase latency), the system combines them so that controllers can gather information and immediately initiate appropriate actions based on that information, while maintaining coordination through inter-controller communication.

Inventive Principle:
Principle #5Merging (Combining)

2Productivity

If multiple storage array controllers operate independently, then responsiveness and real-time data processing improve, but system coordination and unified management become more difficult

Engineering Contradiction:
Improvereal-time data processing capabilityVSAvoidsystem coordination complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent implements feedback mechanisms where storage array controllers continuously exchange information about system state, actions taken, and performance metrics. This feedback loop enables independent controllers to maintain real-time responsiveness while automatically coordinating their actions based on shared system information, reducing the complexity of manual coordination.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent creates universal communication protocols and data structures that enable multiple independent controllers to interact through standardized interfaces. This universality allows controllers to operate independently while maintaining seamless coordination, as each controller can understand and respond to information from any other controller using the same standardized framework.

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

Data Source

PatentUS11032123B1Hierarchical storage system management
Publication Date: 2021.06.08 PURE STORAGE INC
  • US11032123B1 patent drawing
  • US11032123B1 patent drawing
  • US11032123B1 patent drawing

AI summary

Distributing management responsibilities for a storage system that includes a storage array controller and a plurality of storage devices, including: identifying a plurality of elements in the storage system; for each of the plurality of elements in the storage system, creating a distributed manager, wherein each distributed manager is configured for gathering information describing the state of the associated element in the storage system, determining an action to perform against the associated element in the storage system, and executing an approved action against the associated element in the storage system; and creating a distributed management hierarchy that includes each of the distributed managers.