NVDIMM Save Controller Thermal Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In information handling systems, the simultaneous power and thermal management during save operations in Non-volatile Dual In-line Memory Modules (NVDIMMs) is challenging, as ensuring sufficient power does not guarantee persistence of information due to excessive heat generation, especially during power failures in data centers with multiple server boards.
Innovation Solution
Implementing a programmable transfer rate for data from volatile to non-volatile memory during save operations, controlled by a NVM save controller, which adjusts the data transfer rate based on thermal sensors to maintain thermal stability and reduce power consumption, and staggering save operations across multiple NVDIMMs to manage thermal output and power usage effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data transfer rate is increased during save operations, then productivity is improved, but temperature increases causing thermal management issues
Solution Approach 1:
The patent implements dynamic adjustment of data transfer rate based on real-time thermal conditions. The system monitors temperature during save operations and dynamically modifies the transfer rate to maintain thermal stability, allowing higher rates when cool and reducing rates when thermal thresholds are approached.
Solution Approach 2:
The system changes operational parameters (data transfer rate) in response to thermal conditions. By adjusting this parameter based on temperature feedback, the system optimizes the balance between productivity and thermal management during save operations.
2Productivity
If save operations are performed simultaneously across multiple NVDIMMs, then productivity is improved, but power consumption increases
Solution Approach 1:
The patent segments the save operation workload across multiple NVDIMMs with staggered timing. Instead of simultaneous operations, each NVDIMM performs save operations in sequence or with time offsets, dividing the total power demand into manageable portions that prevent power supply overload.
Solution Approach 2:
The system implements periodic save operations across multiple NVDIMMs with controlled intervals. By introducing time periods between operations on different modules, the system maintains throughput while spreading power consumption over time to avoid peak demand issues.
3Temperature
If save operations are delayed to manage power and thermal output, then temperature and power consumption are controlled, but loss of time increases
Solution Approach 1:
The system performs preliminary thermal characterization and power analysis to predict optimal save operation timing. By pre-determining safe operational windows based on thermal and power constraints, the system can schedule save operations to minimize delays while ensuring safety requirements are met.
Solution Approach 2:
The patent implements feedback mechanisms that monitor thermal and power conditions in real-time during save operations. This feedback allows the system to dynamically adjust operation timing and rate, reducing unnecessary delays by stopping only when truly required for thermal or power management reasons.
Data Source
AI summary
An information handling system includes a memory module having a volatile memory, a non-volatile memory, and a save controller configured to execute a save operation that transfers at least all modified information of the volatile memory to the nonvolatile memory. A processor of the information handling system is configured to access the volatile memory of the first memory module. A management controller of the information handling system is configured to, during boot operation of the information handling system send a signal to the first memory module to initiate the save operation of the first memory module, to monitor a first thermal indicator at a location proximate to first memory module during the save operation of the first memory module, and determines a configuration of the information handling system during normal operation based upon whether the thermal indicator exceeds a first threshold.