RAID Group Blocking Determination Using Disk Status Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional RAID systems face challenges in determining whether to block a RAID group due to laborious operations, increased complexity, and lack of shared logic, leading to inefficient maintenance and potential data loss when both access paths become unavailable.
Innovation Solution
A RAID blocking determining method that classifies disks based on their status and access path presence, using a threshold comparison to determine blocking possibility, allowing for shared logic and reduced maintenance complexity, and enabling quick recovery notifications to the host.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional RAID blocking determination method using detailed tables is used, then blocking accuracy is improved, but maintenance complexity and operation labor increase
Solution Approach 1:
The patent changes the determination parameters from detailed event-status combinations to simple disk count thresholds. Instead of consulting complex tables with multiple events and statuses, the system only needs to count broken disks and compare against a threshold value, dramatically simplifying maintenance while preserving blocking determination accuracy.
Solution Approach 2:
The patent extracts the essential blocking determination logic from complex tables into a simple threshold comparison rule. By taking out only the critical parameter (number of broken disks) and comparing it against a predetermined threshold, the system eliminates the need to maintain complex event-status tables while retaining the core blocking determination functionality.
2Measurement precision
If conventional RAID blocking determination method is used, then blocking accuracy is improved, but operation time and productivity decrease
Solution Approach 1:
The patent transforms the operation from consulting complex multi-dimensional tables to simple arithmetic comparison. The system counts broken disks and compares against a threshold, which is a much faster operation than table lookup and complex logic evaluation, thereby improving operation speed while maintaining accuracy.
Solution Approach 2:
The patent pre-calculates and stores the threshold value based on RAID level characteristics before runtime. This preliminary action allows the system to use simple threshold comparison during operation instead of performing complex calculations or table lookups in real-time, significantly improving operation speed.
3Reliability
If conventional RAID blocking determination method is used, then data protection is improved, but host process termination occurs abnormally
Solution Approach 1:
The patent notifies the host in advance that a RAID group will be blocked before the blocking takes effect. This preliminary notification allows the host to prepare for the blocking and avoid abnormal termination of processes, while the RAID group still gets blocked to protect data integrity.
Solution Approach 2:
The patent implements feedback by notifying the host system about upcoming RAID blocking. This feedback mechanism allows the host to adjust its operations accordingly, preventing abnormal process termination while maintaining data protection through the blocking mechanism.
Data Source
AI summary
For each RAID group to be determined for blocking, each disk is classified into three types of totalization unit based on the status of each disk belonging to the RAID group or the presence/absence of an access path to each disk, and the number of disks corresponding to each totalization unit is totalized. The totalization unit is a “used disk”, an “unused disk”, and a “loop-down disk”. The totalization result is compared with the threshold condition set for each RAID level (for example, “unused disk”=0 and “loop-down disk”=‘1 or more’), and the blocking possibility is determined.