Memory Device Configuration Recovery via Pre-stored SPD Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Memory device failures, particularly in the configuration region of DIMM memory devices, lead to un-configurability and effective data loss, impacting system performance and memory bandwidth, with existing technologies lacking effective recovery mechanisms.
Innovation Solution
An Information Handling System (IHS) with a memory device management engine that identifies failures during boot, retrieves and uses previously stored memory device configuration information to configure the memory device, allowing for recovery even if the configuration region is faulty.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If memory device configuration information is stored only in the SPD EEPROM device on the memory device itself, then the memory device can be configured during boot, but a memory cell failure in the SPD EEPROM device renders the configuration information unavailable and prevents configuration
Solution Approach 1:
The system performs preliminary action by capturing and storing memory device configuration information in a database during a previous successful boot operation, before any potential failure occurs. This pre-stored information serves as a backup that can be retrieved when the SPD EEPROM fails, preventing configuration information loss and enabling recovery without requiring manual reconfiguration.
Solution Approach 2:
The patent introduces an intermediary component - a database separate from the SPD EEPROM device - to store configuration information. This intermediary serves as a mediator between the original configuration source and the memory device, providing an alternative path for configuration data when the SPD EEPROM fails, thus resolving the contradiction between configuration availability and information loss risk.
2Ease of repair
If the SPD EEPROM device fails, then the memory device becomes un-configurable, but replacing the entire memory device is costly and unnecessary
Solution Approach 1:
The patent applies segmentation by separating the configuration information storage function from the physical memory device hardware. Instead of treating the SPD EEPROM as an inseparable part of the memory device, the system extracts and stores configuration information independently in a database. This allows the configuration function to be recovered independently of the failed hardware component, enabling repair without replacing the entire expensive memory device.
Solution Approach 2:
The system creates a copy of the configuration information from the failed SPD EEPROM and stores it in a database during normal operation. When failure occurs, this copied information is retrieved and used to reconfigure the memory device, eliminating the need to replace the hardware. This copying approach transforms an unrecoverable hardware failure into a recoverable software/configuration issue.
3Measurement precision
If configuration information is retrieved and stored during each boot operation, then configuration accuracy is maintained, but system boot time increases
Solution Approach 1:
The system performs the configuration information capture and storage action preliminarily during a previous successful boot operation. This pre-captured information is then reused in subsequent boots without requiring re-retrieval from the failed SPD EEPROM, thereby maintaining configuration accuracy while eliminating the time penalty of repeated retrieval attempts during boot operations.
Data Source
AI summary
A memory device failure recovery system includes a memory device management engine that is coupled to a first memory device via a first memory device slot, and a memory device management database. The memory device management engine identifies that the first memory device has experienced a failure in a configuration region of the first memory device during a current boot operation and, in response, retrieves memory device component information and memory device configuration information that is stored in the memory device management database and that was retrieved as part of a prior boot operation from a memory device that was connected to the first memory device slot. During the current boot operation, the memory device management engine determines whether first memory device components on the first memory device correspond to the memory device component information and, if so, uses the memory device configuration information to configure the first memory device.