Evacuating Object Components from Failing Disks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed storage systems, a failing disk drive can cause data loss and accessibility issues due to its slow performance, leading to the locking up of storage objects and potential data loss if not addressed promptly.
Innovation Solution
A method is implemented to detect failing disk drives by monitoring performance metrics and designating them as failing. An object coordinator marks object components with an ‘evacuate’ flag for migration to other disk drives, ensuring data accessibility, while components not affecting accessibility are marked with an ‘absent’ flag with lower priority or not migrated, thereby preventing data loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If object components are migrated from failing disk drives, then data accessibility is improved, but system complexity increases due to evacuation flags and coordination mechanisms
Solution Approach 1:
The system performs preliminary actions by detecting slow disk drives before they completely fail and proactively evacuating object components. The object coordinator identifies failing disks through performance monitoring and initiates evacuation of critical components before total failure occurs, preventing data loss and maintaining accessibility.
Solution Approach 2:
The object coordinator acts as an intermediary between disk drives and storage objects. It receives notifications about failing disks, determines which object components are critical, and coordinates their evacuation to new locations. This intermediary manages the complexity of tracking component locations and dependencies.
2Reliability
If all object components are migrated from failing disk drives, then data loss prevention is improved, but migration time and system performance deteriorate
Solution Approach 1:
The system applies different evacuation priorities to different object components based on their criticality. Critical components that would cause inaccessibility are evacuated with high priority, while non-critical components are evacuated with lower priority or not immediately. This selective approach based on local quality assessment optimizes migration time while preventing data loss.
Solution Approach 2:
The system performs partial evacuation by focusing resources on evacuating only the critical object components necessary to maintain accessibility, rather than evacuating all components equally. This partial action on the most important components achieves data loss prevention with reduced migration time compared to complete evacuation.
3Reliability
If performance metrics are continuously monitored to detect failing disk drives, then reliability is improved, but energy consumption and system overhead increase
Solution Approach 1:
The system uses periodic monitoring of disk drive performance metrics rather than continuous monitoring. The object coordinator and evacuating agent check disk performance at intervals, tracking metrics such as read/write speeds and error rates. This periodic action detects failing disks reliably while reducing energy consumption and system overhead compared to continuous monitoring.
Data Source
AI summary
Exemplary methods, apparatuses, and systems determine that a disk drive in a distributed storage system is failing. In response to determining that the disk drive is failing, it is determined if an object has object components stored on the failing disk drive. Object components are flagged based on whether the object components will become inaccessible in the event the disk drive fails or is unmounted. Object components that are flagged with an evacuate flag are migrated from the failing disk drive to one or more other disk drives in the distributed storage system to prevent objects from becoming inaccessible.