Persistent Memory Timing Characterization for NVDIMM Energy Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information handling systems using Non-Volatile Dual In-line Memory Modules (NVDIMMs face impractical and cost-ineffective challenges in designing energy storage devices to support worst-case timing across various configurations, as they require tens of seconds to minutes to copy data from DRAM to flash memory during power loss.
Innovation Solution
An information handling system that includes a processor and a management controller to determine the energy required for a save operation from volatile to non-volatile memory and assess whether available energy exceeds this requirement, allowing the system to decide whether to support persistent memory capabilities, potentially reconfiguring to reduce energy needs if necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the energy storage device is designed to support worst-case timing across all NVDIMM configurations, then the persistent memory reliability is improved, but the device complexity and cost increase significantly
Solution Approach 1:
The patent implements dynamic timing characterization that adapts to specific NVDIMM configurations rather than using fixed worst-case timing. The system measures actual save operation times for different configurations (capacity, organization, error correction modes, encryption settings, wear leveling states) and adjusts the energy storage device sizing accordingly. This dynamic approach allows the energy storage device to be optimized for the actual workload and configuration rather than being oversized for all possible scenarios, thereby reducing complexity while maintaining reliability.
Solution Approach 2:
The patent changes the parameter of timing characterization from static worst-case to dynamic configuration-specific measurements. By measuring save operation times under different operational parameters (error correction code modes, encryption settings, state-of-wear leveling), the system determines the actual energy requirements for each configuration. This allows the energy storage device to be designed with precise capacity matching the specific NVDIMM configuration, avoiding the need to support all possible worst-case scenarios simultaneously.
2Reliability
If the energy storage device is designed to support worst-case timing across all NVDIMM configurations, then the persistent memory reliability is improved, but the cost becomes impractical
Solution Approach 1:
The system dynamically characterizes timing based on actual NVDIMM configurations rather than using static worst-case assumptions. By measuring save operation times for specific configurations (capacity, organization, error correction modes, encryption settings, wear leveling states), the patent enables cost-effective energy storage device sizing that matches actual requirements. This eliminates the need to overspecify energy storage capacity to cover all possible configurations, thereby reducing cost while maintaining reliability for the deployed configuration.
Solution Approach 2:
The system performs self-characterization of timing requirements by measuring actual save operation times during operation. The NVDIMM and information handling system work together to determine the specific energy requirements through empirical measurement rather than relying on conservative manufacturer specifications. This self-service approach allows the system to optimize energy storage device selection based on actual performance characteristics, reducing unnecessary costs associated with worst-case design margins.
3Productivity
If the save operation time is reduced for specific configurations, then the productivity is improved, but the adaptability across different NVDIMM configurations decreases
Solution Approach 1:
The patent implements dynamic timing characterization that adapts to each NVDIMM configuration rather than using a single fixed timing value. The system measures save operation times for different configurations (capacity, organization, error correction modes, encryption settings, wear leveling states) and determines configuration-specific timing values. This allows the system to achieve optimal save operation speed for each specific configuration while maintaining adaptability across all configurations through the measurement and characterization process.
Solution Approach 2:
The patent changes the timing parameter from a single fixed value to multiple configuration-specific values. By measuring and characterizing save operation times under different operational parameters, the system enables productivity optimization for each configuration while maintaining adaptability through the comprehensive characterization data. The system can select appropriate timing values based on the actual configuration, achieving both speed optimization and broad adaptability.
Data Source
AI summary
In accordance with embodiments of the present disclosure, an information handling system may include a processor and a management controller communicatively coupled to the processor and configured to, during boot of the information handling system determine a first amount of energy required by the information handling system to perform a save operation to transfer data from a volatile memory to a non-volatile memory of a persistent memory in response to a loss of power for supplying electrical energy to the information handling system, determine whether a second amount of energy available for providing electrical energy for the save operation in response to the loss of power exceeds the first amount of energy, and responsive to determining whether the second amount of energy exceeds the first amount of energy, determine whether to support the persistent memory.