RAID Firmware Activation via Disk Weighted Index
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Firmware activation in RAID storage environments is inefficient due to prolonged activation times, unawareness of subsystem load during updates, and sequential activation methods, which impact I/O performance.
Innovation Solution
A method that involves obtaining I/O metrics for each disk, calculating a disk weighted index (DWI) to determine the load, comparing it to thresholds, and triggering firmware activation based on aggregated metrics and counters to optimize the activation process, allowing for parallel activation in certain cases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If firmware activation is performed for all disks in a RAID configuration, then firmware update completeness is improved, but activation time and subsystem load increase
Solution Approach 1:
The patent segments disks into multiple groups based on their I/O metrics and activation requirements. Instead of activating all disks simultaneously, the system divides them into groups that can be activated in parallel, reducing total activation time while maintaining complete firmware updates. This is achieved by evaluating I/O metrics for each disk and creating groups that minimize sequential dependencies.
Solution Approach 2:
The patent dynamically adjusts the activation process based on real-time I/O metrics and system conditions. The system continuously monitors disk I/O performance and adapts the activation strategy accordingly, switching between sequential and parallel activation modes based on current load conditions. This dynamic approach optimizes the balance between update completeness and activation time.
2Device complexity
If sequential activation is used for disks in a RAID subsystem, then activation process simplicity is improved, but total activation time increases
Solution Approach 1:
The patent segments the activation process into multiple parallel tracks based on disk I/O metrics. Disks are divided into groups that can be activated simultaneously rather than sequentially, reducing total activation time. The system maintains manageable complexity by establishing clear grouping criteria based on I/O performance thresholds.
Solution Approach 2:
The patent implements periodic monitoring and evaluation of I/O metrics during the activation process. The system periodically checks disk performance and adjusts activation timing accordingly, allowing for optimized parallel activation while maintaining process manageability. This periodic action enables the system to balance simplicity with time efficiency.
3Productivity
If firmware activation is triggered without considering subsystem load, then update speed is improved, but I/O performance impact increases
Solution Approach 1:
The patent implements a feedback mechanism that continuously monitors disk I/O metrics and uses this information to guide the activation process. The system evaluates I/O performance data before and during activation, adjusting the activation strategy to minimize impact on ongoing I/O operations. This feedback loop ensures that updates are scheduled when they will cause minimal disruption to storage performance.
Solution Approach 2:
The patent dynamically adjusts activation timing and grouping based on real-time I/O load conditions. When subsystem load is high, the system delays activation or groups disks differently to minimize I/O impact. When load is low, activation proceeds more aggressively. This dynamic adaptation optimizes the balance between update speed and performance impact.
Data Source
AI summary
Methods, systems, and computer programs encoded on computer storage medium, for performing firmware activation, including obtaining input/output (I/O) metrics of the disks; determining an index value for the disk based on a disk weighted index (DWI); determining a DWI value for the array of disks based on the index value for each disk of the array of disks; comparing the DWI value for the array of disks to a DWI threshold; determining that the DWI value for a first array of disks is greater than the DWI threshold, and in response: aggregating the I/O metrics for each disk of the first array of disks; comparing the aggregated I/O metrics for the first array of disks to an activation threshold; determining that the aggregated I/O metrics for the first array of disks is less than the activation threshold, and in response triggering firmware activation for the first array of disks.