Automated SCM Volume Recreation via NvRAM Metadata Backup
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Storage Class Memory (SCM) devices, which are intermediate storage solutions, can lose persistence, resulting in the erasure of data and volumes, requiring manual recreation and reassignment, a time-consuming process.
Innovation Solution
A method where volume metadata is stored in a less volatile storage device, such as Non-Volatile Random Access Memory (NvRAM), allowing for automatic recreation of lost volumes by retrieving and reconfiguring the metadata upon persistence loss, eliminating the need for manual recreation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If volume metadata is stored only in SCM storage device, then storage performance is optimized, but data loss occurs upon persistence loss requiring manual recreation
Solution Approach 1:
The system performs preliminary action by storing volume metadata in advance in a less volatile storage device before any persistence loss occurs. This advance preparation ensures that when persistence loss happens at the SCM storage device, the metadata is already available for automatic recreation without manual intervention, thus resolving the contradiction between optimization for performance and assurance of reliability.
2Reliability
If manual recreation of volumes is performed after persistence loss, then data integrity can be restored, but significant time is consumed
Solution Approach 1:
The system implements self-service by automatically detecting persistence loss, retrieving stored metadata, and recreating volumes without requiring manual user intervention. The automation process includes receiving a notification of persistence loss, retrieving the volume metadata from less volatile storage, and automatically reconfiguring the storage volume by submitting the metadata to the SCM device driver, thereby eliminating time-consuming manual operations while ensuring data integrity restoration.
Solution Approach 2:
The system uses feedback by implementing a notification mechanism that detects persistence loss events and triggers the automatic metadata retrieval and volume recreation process. This feedback loop ensures that as soon as persistence loss is detected, the system automatically initiates restoration procedures using the pre-stored metadata, minimizing downtime while maintaining data integrity.
3Extent of automation
If volume metadata is stored in less volatile storage device, then automatic recreation is enabled, but additional storage resources are required
Solution Approach 1:
The system uses an intermediary approach by introducing a less volatile storage device as a mediator between the SCM storage device and the system managing metadata. This intermediary storage component enables automatic volume recreation by holding backup metadata, thus achieving high automation while using additional storage resources in a controlled and necessary manner.
Data Source
AI summary
A computing device, a non-transitory computer readable storage medium, and a method are provided for automatically recreating lost storage volumes. A volume command including volume metadata associated with a storage volume of an SCM storage device is received. The volume metadata is stored at another storage device that is less volatile than the SCM storage device. The storage volume is configured in accordance with the volume command including forwarding the volume command to an SCM device driver. A notification of a persistence loss at the SCM storage device is received subsequent to forwarding the volume command. The volume metadata is retrieved from the other storage device subsequent to receiving notification of the persistence loss. The storage volume is automatically re-configured in accordance with the volume command including submitting the retrieved volume metadata to the SCM device driver.


