Redundant I/O Queue De-allocation in Multi-site Storage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Multi-site storage systems face inefficiencies due to redundant Input/Output (I/O) queues being maintained across multiple storage sites, which waste computing resources and hinder system performance.

Innovation Solution

A computer-implemented method identifies redundant I/O queues in a multi-site storage system, notifies the host to terminate connections, and deallocates memory allocated for processing these queues, thereby optimizing resource utilization and enhancing system efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If I/O queues are maintained at both primary and backup storage systems, then data redundancy and disaster recovery capability are improved, but system resource utilization deteriorates due to redundant queue processing

Engineering Contradiction:
Improvedisaster recovery capabilityVSAvoidsystem resource utilization
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system dynamically adjusts I/O queue configuration based on operational state. During normal operation, queues are maintained at both primary and backup systems for redundancy. Upon detecting primary system failure, the system dynamically reconfigures to activate backup queues, eliminating the need to permanently maintain duplicate queue structures and their associated resource overhead.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

Instead of maintaining full functional copies of I/O queues at the backup system, the invention uses queue information copying and registration mechanisms. The backup system receives and stores queue registration information from the primary system, allowing it to assume queue processing roles only when needed, rather than continuously maintaining active queue structures.

Inventive Principle:
Principle #26Copying

2Productivity

If memory is allocated for processing I/O queues at the backup storage system, then queue processing capability is improved, but memory resource waste increases due to redundant queue allocation

Engineering Contradiction:
Improvequeue processing capabilityVSAvoidmemory resource waste
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

The backup system performs preliminary actions by receiving and storing queue registration information from the primary system in advance. This allows the backup system to have queue metadata available before actual queue activation is needed, so that when failover occurs, memory resources can be rapidly allocated without wasting resources on continuous maintenance of full queue structures.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements dynamic memory allocation where backup queues are allocated only when needed (upon primary system failure detection). When the primary system is operational, the backup system discards or deallocates memory resources associated with redundant queue structures, recovering those resources for other uses while maintaining the ability to quickly reactivate queues when needed.

Inventive Principle:
Principle #34Discarding and recovering

Data Source

PatentUS20210263676A1Queue management in multi-site storage systems
Publication Date: 2021.08.26 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US20210263676A1 patent drawing
  • US20210263676A1 patent drawing
  • US20210263676A1 patent drawing

AI summary

A computer-implemented method to identify redundant Input/Output (I/O) queues in a multi-site storage system. The method includes receiving, from a host, by a backup storage system, a request to process a first set of Input/Output (I/O) queues, wherein the backup storage system is a second subsystem in a multi-site storage system. The method includes, allocating memory on the backup storage system. The method includes, identifying a second set of I/O queues established at a primary storage system, a first subsystem in the multi-site storage system. The method includes, determining the first set of I/O queues and the second set of I/O queues are redundant. The method includes, responsive to determining queues are redundant: notifying via the host, that the first set of the redundancy, terminating a connection between the host and the backup storage system, and de-allocating the memory to process the first set of I/O queues.