Recover State I/O for RAID Degraded Storage Recovery

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current RAID technologies face data loss issues due to hardware failures, incomplete write operations, and degraded states, particularly when storage devices recover from failures with different power-on times and constraints on debounce times, leading to inconsistent data and potential data loss.

Innovation Solution

The introduction of Recover State I/O operations, including Recover State Read and Write, which allow the RAID group to return a 'RECOVER NEEDED' message, ensuring no media modifications are made during recovery, and involve splitting I/O requests to handle degraded states and inconsistent data, thereby preventing data loss.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If WriteBack technology is used to improve I/O performance, then data transmission rate is improved and access delay is reduced, but data loss occurs when RAID group recovers from hardware failure

Engineering Contradiction:

Improvedata transmission rateVSAvoiddata loss during recovery

Core Design Contradiction:

SpeedVSReliability

Solution Approach 1:

The system performs preliminary actions by detecting degraded state before I/O operations and pre-establishing recovery protocols. When the RAID group is in a degraded state, the system identifies this condition in advance and implements special handling mechanisms to prevent data loss during the recovery process.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms by continuously monitoring the health status of the RAID group and adjusting I/O operations accordingly. When degradation is detected, the system receives feedback about the degraded state and modifies its behavior to avoid incomplete write operations that could cause data loss during recovery.

Inventive Principle:

Principle #23Feedback

2Reliability

If storage device recovers from hardware failure, then system availability is improved, but data consistency becomes inconsistent due to different power-on times

Engineering Contradiction:

Improvesystem availabilityVSAvoiddata consistency

Core Design Contradiction:

ReliabilityVSStability of the object's composition

Solution Approach 1:

The system segments the recovery process into distinct phases: detection phase, stabilization phase, and normal operation phase. During the stabilization phase, the system separates I/O operations into two types: normal I/O for non-degraded devices and recover state I/O for degraded devices, ensuring data consistency while maintaining availability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system dynamically adjusts its behavior based on the recovery state of storage devices. When devices are recovering, the system transitions from normal I/O processing to recover state I/O processing, and when recovery is complete, it returns to normal operation. This dynamic adaptation ensures data consistency throughout the recovery process.

Inventive Principle:

Principle #15Dynamics

3Ease of operation

If I/O request is sent to degraded storage device, then data access is maintained, but incomplete write operations cause data loss

Engineering Contradiction:

Improvedata access continuityVSAvoiddata loss from incomplete writes

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The system introduces an intermediary mechanism - the recover state I/O processing logic - that mediates between normal I/O operations and the degraded storage state. This intermediary layer detects degraded states and transforms normal I/O requests into recover state I/O requests, which then trigger appropriate recovery handling to prevent incomplete writes.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system applies preliminary anti-action by preventing incomplete write operations before they can cause data loss. When a degraded state is detected, the system preemptively stops normal write operations and initiates recovery procedures, counteracting the potential harmful effect of incomplete writes before they occur.

Inventive Principle:

Principle #9Preliminary anti-action

4Device complexity

If cache device uses normal I/O request during degraded state, then processing simplicity is maintained, but data loss window increases

Engineering Contradiction:

Improveprocessing simplicityVSAvoiddata loss window

Core Design Contradiction:

Device complexityVSLoss of time

Solution Approach 1:

The system applies local quality by treating different I/O operations differently based on their characteristics. Normal I/O requests continue to be processed normally for non-degraded devices, while recover state I/O requests are specially handled for degraded devices. This localized differentiation reduces the data loss window without significantly increasing overall processing complexity.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS10983874B2Processing a recover state input/output request

Publication Date: 2021.04.20 EMC IP HLDG CO LLC

AI summary

A storage system comprises a cache device and a storage device group. A technique involves: the cache device of the storage system receiving from a host an input/output (I/O) request for the storage system; and upon determining that the storage system is just recovered from a hardware failure and the storage device group in the storage system is in a ready but degraded state, safely performing data write/read by sending a new Recover State I/O request to the storage device group. Such a technique may be used to reduce a Data Lost window, reduce or avoid data loss. In addition, such a technique may be implemented in a current RAID technology, namely, can be used in combination with the current RAID technology.