Auxiliary Storage Arrays for Primary Site Data Recovery

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing storage network recovery techniques are inefficient and inconvenient due to high latency and performance issues when a primary storage array goes offline or experiences errors, requiring the copying of entire data pools from secondary sites, which can be time-consuming and risk-prone.

Innovation Solution

A method and system that detect errors in primary storage devices and request subsets of mirrored data from a backup system, storing them in auxiliary storage arrays co-located with primary arrays, allowing for efficient data recovery and reducing rebuild time by replacing faulty devices with corresponding auxiliary devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the entire extent data pool is copied from secondary site to primary site for recovery, then data recovery is achieved, but latency increases and recovery time is extended

Engineering Contradiction:
Improvedata recovery capabilityVSAvoidrecovery latency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent divides the recovery process into segments by copying only the specific data subset needed for recovery rather than the entire extent data pool. The management module identifies and copies only the portion of data corresponding to the offline array, segmenting the large-scale data transfer into a targeted, smaller operation that reduces latency while maintaining full recovery capability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts and copies only the necessary subset of data from the secondary site backup storage devices to the primary site auxiliary storage devices. Instead of transferring the complete data pool, the system identifies and extracts specifically the data corresponding to the failed array, significantly reducing transfer time and latency while ensuring the required data is available for recovery.

Inventive Principle:
Principle #2Taking out (Extraction)

2Productivity

If multiple devices are rebuilt simultaneously during an overlapping period, then recovery parallelism is improved, but storage network performance deteriorates

Engineering Contradiction:
Improverecovery parallelismVSAvoidstorage network performance
Core Design Contradiction:
ProductivityVSPower

Solution Approach 1:

The patent introduces auxiliary storage devices at the primary site that are pre-configured and ready to receive data subsets from the secondary site. These auxiliary devices serve as intermediate buffers, allowing rebuild operations to be staged and managed sequentially or in controlled parallel batches, preventing overwhelming the storage network while maintaining high overall recovery productivity through the pre-prepared auxiliary infrastructure.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS10877860B2Copying data from mirrored storage to auxiliary storage arrays co-located with primary storage arrays
Publication Date: 2020.12.29 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US10877860B2 patent drawing
  • US10877860B2 patent drawing
  • US10877860B2 patent drawing

AI summary

Methods that copy data from mirrored storage to auxiliary storage arrays co-located with primary storage arrays are provided. One method includes requesting a subset of the data from a backup system mirroring the set of data at a remote location in response to detecting an error in a storage device of an array of primary storage devices storing a set of data. The method further includes receiving the subset of the data from the backup system and storing the subset of the data in an array of auxiliary storage devices co-located with the array of primary storage devices in which the subset of the data can correspond to data stored on the storage device. Systems and computer program products for performing the above method are also provided.