RAID Spare Capture via SCSI Enclosure Services Processor
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In high-density packaging storage systems, existing methods for replacing or migrating a RAID array member consume significant system resources, impacting performance during data migration and repair activities, especially when dealing with mixed storage device configurations and multiple devices within a single service boundary.
Innovation Solution
The method involves using a RAID controller to capture a spare storage device and initialize a SCSI enclosure services processor (SESP) to read data from the RAID array member and transfer it to the spare device, both residing within a SAS expander local port domain, minimizing resource usage and optimizing data migration within a high-density packaging configuration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a RAID controller performs mirror rebuild to synchronize data between source and target devices, then data migration is achieved, but significant RAID controller and storage device fabric resources are consumed, impacting overall system performance
Solution Approach 1:
The patent extracts the data copying function from the RAID controller to a dedicated SES processor. The SES processor is initialized by the RAID controller to perform the actual data read from source and write to target devices, while the RAID controller retains only the coordination and management functions. This separation removes the resource-intensive copying workload from the RAID controller, resolving the contradiction between complete data migration and maintained system performance.
Solution Approach 2:
The SES processor acts as an intermediary between the RAID controller and storage devices for the data copying operation. The RAID controller initializes and manages the SES processor, which then directly handles the data transfer between source and target devices through the storage fabric. This intermediary approach enables data migration without burdening the RAID controller with heavy copying tasks, thus maintaining system performance while ensuring reliable data migration.
2Ease of manufacture
If data migration is performed using existing storage devices and network fabrics, then cost-effectiveness is improved, but system resources are still significantly consumed during the migration activity
Solution Approach 1:
The SES processor is a dedicated component already present in the storage system that can be initialized to perform data copying tasks. Rather than requiring additional hardware or external resources, the system uses its existing SES processor to handle the migration workload. This self-service approach enables cost-effective migration using existing resources while the dedicated nature of the SES processor ensures it can handle the workload efficiently without excessive resource consumption.
3Adaptability or versatility
If RAID array repair action is performed with mixed storage device types, then device replacement flexibility is improved, but configuration optimality deteriorates due to intermix of storage device types with mismatched characteristics
Solution Approach 1:
The patent segments the RAID array into individual replaceable devices within FRUs. Each storage device can be independently identified, captured, and replaced without affecting the entire array. This segmentation allows flexible device replacement while maintaining array functionality, and the data migration process can selectively handle each device type mismatch scenario, preserving configuration optimality for the remaining devices.
Solution Approach 2:
The system handles mixed device types by changing parameters during the migration process. When a device is replaced with a different type, the SES processor adjusts the data copying parameters to accommodate the new device's characteristics (capacity, speed, interface rate). This parameter adaptation enables flexible replacement while maintaining optimal configuration for the overall array by appropriately managing the migrated data on the new device type.
Data Source
AI summary
A computer-implemented method replaces a redundant array of independent disks (RAID) array member storage device configured in high density packaging. The high density packaging includes a plurality of field-replaceable units (FRUs). Each FRU has an associated plurality of storage devices. A first spare storage device of a first FRU is captured. The first FRU is dedicated as a plurality of spare storage devices by a RAID controller. A small computer system interface (SCSI) enclosure services (SES) processor (SESP) is initialized to read data from the RAID array member storage device and transfer the data to the first spare storage device. The first spare storage device and the RAID array member storage device are co-resident within a serial attached small computer system interface (SAS) expander local port domain.