Storage Disk File Subsystem Defect Management via Workload Migration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In IT environments, managing workloads on IT resources with critical components is challenging due to the difficulty in detecting defects until they cause problems, which can lead to issues like data corruption, component failure, or environmental hazards.
Innovation Solution
A system and method for workload and defect management that involves communicating identifiers of IT resources, receiving maintenance information, and implementing policies to migrate workloads or storage disk files from defective components to healthy ones, using a systems management server and an engineering change advisory (ECA) database to detect and respond to defects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If proactive defect detection and workload migration are implemented, then reliability and availability are improved, but device complexity and management overhead increase
Solution Approach 1:
The patent introduces a defect management system as an intermediary layer between storage devices and workload management systems. This system receives defect notifications from manufacturers, stores them in a database, and automatically triggers workload migration policies, thereby resolving the complexity of direct proactive detection while improving reliability
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring defect notifications from storage device manufacturers and automatically adjusting workload placement based on received defect information. This closed-loop feedback enables proactive reliability improvement without requiring complex direct detection systems
2Speed
If automatic workload migration is implemented, then response time to defects is improved, but loss of time for data migration and system reconfiguration occurs
Solution Approach 1:
The patent pre-configures migration policies and identifies alternative storage resources before defects occur. When defects are detected, the system already has predetermined migration paths ready, enabling rapid automatic migration without time-consuming decision-making or system reconfiguration
Solution Approach 2:
The system creates virtual copies of workloads that can be quickly instantiated on alternative storage resources. This copying approach enables fast failover by deploying replica workloads rather than migrating entire systems, reducing the time loss associated with automatic migration
3Difficulty of detecting and measuring
If continuous monitoring of storage devices is implemented, then defect detection capability is improved, but use of energy and computational resources increases
Solution Approach 1:
The patent leverages the storage device manufacturer's existing monitoring and notification systems, which continuously track device health and defect conditions. This self-service approach allows the system to benefit from continuous monitoring without requiring additional energy-intensive monitoring infrastructure, as the storage devices themselves provide the monitoring function
Data Source
AI summary
In accordance with one or more embodiments of the present invention, methods and systems disclosed herein provide for storage disk file and defect management. According to an aspect, a method includes communicating, to a first computing device, an identifier associated with a storage device. The method also includes receiving, from the first computing device, maintenance information associated with the identifier. Further, the method also includes implementing a storage disk file management policy at the storage device based on the maintenance information. Maintenance or defect information may be provided, for example, for a disk subsystem comprised within a storage device. A disk subsystem with an identified potential defect, even though potential defect is latent, creates a risk to the software workloads running on the IT resource and a potential hazard in a datacenter.