Storage System Path Failure Isolation via Block Range Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In complex systems where multiple modules are coupled through multiple paths for different applications, a failure in one path can inadvertently affect other paths, hindering normal operations and CPU functionality.

Innovation Solution

A computer system with a control unit and storage unit, where each module is coupled via multiple paths with distinct applications, utilizing block range information to isolate and block the affected path and its corresponding module, ensuring minimal disruption during failures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a failure occurs on one of the paths and only the failed path is blocked, then the other paths can continue operating, but the failure may spread to other paths and hinder normal operations

Engineering Contradiction:
Improvepath operation continuityVSAvoidfailure propagation
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The system segments the failure isolation scope by path, maintaining independent blocking control for each path coupled to a module. When a failure occurs on one path, only that specific path is blocked while other paths remain operational, preventing failure propagation while preserving system availability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The blocking mechanism applies local quality by differentiating the blocking state between paths. Each path can be independently blocked or unblocked based on its specific failure status, allowing the system to maintain different operational states for different paths connected to the same module.

Inventive Principle:
Principle #3Local quality

2Reliability

If the module and all coupling paths are blocked when a failure occurs, then failure tolerance is improved, but system availability decreases

Engineering Contradiction:
Improvefailure toleranceVSAvoidsystem availability
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system divides the failure isolation scope into path-level segments rather than blocking the entire module. This segmentation allows selective blocking of only the failed path while keeping the module and other paths operational, thereby maintaining system availability while ensuring failure tolerance.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Instead of blocking all paths connected to a module (excessive action), the system applies partial blocking only to the specific failed path. This partial action is sufficient to contain the failure while avoiding unnecessary disruption to other operational paths, thus preserving system productivity.

Inventive Principle:
Principle #16Partial or excessive action

3Adaptability or versatility

If a single device is coupled by multiple paths for different applications, then versatility is improved, but failure isolation becomes more complex

Engineering Contradiction:
Improvemulti-application supportVSAvoidfailure management complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system segments failure management by path, with each path having independent blocking control. This segmentation simplifies failure management in multi-application scenarios by allowing each path to be managed independently based on its specific failure status, rather than requiring complex coordinated management of all paths.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system establishes preliminary blocking control mechanisms for each path before failures occur. This preliminary setup of independent path-level blocking controls simplifies failure management when failures actually occur, as the system can immediately apply blocking to the specific failed path without complex decision-making.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS10353613B2Computer system and control method therefor for handling path failure
Publication Date: 2019.07.16 HITACHI VANTARA LTD
  • US10353613B2 patent drawing
  • US10353613B2 patent drawing
  • US10353613B2 patent drawing

AI summary

When mounting hardware which is coupled to another portion by a plurality of paths with different applications, despite the hardware being a single device, and a failure occurs in any of the paths, there is a risk that the failure may propagate to other components unless the other paths are also blocked.In order to solve the problem described above, in a storage apparatus to which a device coupled by a plurality of coupling paths with different applications can be mounted, the present invention determines a block range at the time of an occurrence of a failure to be a device and a plurality of coupling paths coupled to the device, manages the block range, and upon an occurrence of a failure, executes failure handling which involves blocking an appropriate block range determined in advance by referring to the information.