Memory Sub-System Composite Temperature for Thermal Throttling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory sub-systems rely on single temperature sensors that fail to accurately measure temperature variations across the system, leading to inadequate thermal protection and inefficient thermal management.
Innovation Solution
Implement a composite temperature calculation method that aggregates device temperature values from multiple sensors within the memory sub-system, using a composite temperature threshold ratio to determine a normalized value representing the overall system temperature, ensuring accurate thermal management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single temperature sensor is used to measure memory sub-system temperature, then the device complexity is reduced, but the measurement precision is insufficient to accurately represent temperature variations across the system
Solution Approach 1:
The patent divides the temperature measurement function into multiple segments by deploying several temperature sensors at different locations within the memory sub-system. Each sensor measures temperature at its specific location, and the memory controller aggregates these distributed measurements to form a comprehensive view of the system's thermal state, thereby improving overall measurement precision without requiring a single complex sensor.
Solution Approach 2:
The patent combines multiple temperature sensor readings into a single aggregated temperature value that represents the overall thermal state of the memory sub-system. The memory controller merges the data from multiple sensors using algorithms that consider the thermal characteristics and locations of each sensor, producing a unified temperature metric for thermal management decisions.
2Measurement precision
If multiple temperature sensors are deployed across the memory sub-system, then the measurement precision improves, but the device complexity increases
Solution Approach 1:
The memory controller is designed to perform multiple functions: it manages memory operations and simultaneously handles thermal management by aggregating data from multiple temperature sensors. This multi-functionality eliminates the need for a separate dedicated thermal management hardware system, thereby improving measurement precision through multiple sensors without proportionally increasing device complexity.
Solution Approach 2:
The memory sub-system performs self-thermal-management by having the memory controller automatically aggregate temperature sensor data and make thermal management decisions without external intervention. The system uses its own internal resources (the memory controller) to process temperature data and execute thermal policies, reducing the need for additional external monitoring and control components.
3Reliability
If aggregated temperature data from multiple devices is used to determine composite temperature, then the reliability of thermal management improves, but the computational complexity increases
Solution Approach 1:
The patent transforms multiple temperature parameters from different sensors into a single composite temperature parameter through mathematical aggregation. The memory controller applies algorithms that weight and combine individual sensor readings based on their locations and thermal significance, changing the parameter representation from multiple discrete values to a unified composite value that drives thermal management decisions.
Solution Approach 2:
The system implements a feedback mechanism where the aggregated temperature data continuously informs thermal management decisions, which in turn affect the operating conditions and temperatures of the memory devices. The memory controller monitors temperature sensors, aggregates the data, and adjusts operational parameters accordingly, creating a closed-loop feedback system that improves reliability through continuous adaptation.
Data Source
AI summary
A plurality of device temperature values that are each indicative of a temperature at a respective device of a plurality of devices of a system is identified. A respective composite temperature threshold ratio is determined for each device of the plurality of devices. A respective normalization value based on the respective composite temperature threshold ratio and the respective device temperature value is determined for each device of the plurality of devices. A largest normalization value of the plurality of devices is determined. A composite temperature of the system based on the largest normalization value of the plurality of devices is set.