Thermal Wear Leveling for IC Die Regions Using Persistent Heat History
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing integrated circuit systems rely on short-term temperature monitoring for thermal control, which fails to address long-term thermally accelerated faults such as negative biased temperature instability and electromigration, leading to wear-out issues among components in 3D packages.
Innovation Solution
Implementing thermal wear spreading logic that uses cumulative temperature data stored in persistent memory to control task execution and memory access operations, spreading thermal wear among die regions based on their wear-out levels to prevent excessive heat buildup.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If short-term temperature monitoring is used for thermal control, then immediate thermal management is achieved, but long-term thermal wear and component degradation are not addressed
Solution Approach 1:
The system performs preliminary action by accumulating temperature data over the component's operational life before wear-out occurs. The persistent memory stores historical temperature information in advance, enabling proactive thermal wear leveling decisions rather than reactive responses to immediate overheating conditions.
Solution Approach 2:
The system dynamically adapts task execution based on accumulated thermal wear data. The thermal wear leveling controller continuously updates task scheduling decisions according to the historical temperature profiles of different die regions, making the system flexible and adaptive to long-term thermal patterns rather than static short-term conditions.
2Productivity
If tasks are concentrated on specific die regions for optimized performance, then processing efficiency is improved, but thermal wear becomes uneven and accelerates component degradation
Solution Approach 1:
The system applies local quality by treating different die regions with different task allocation strategies based on their individual thermal wear characteristics. Each die region's historical temperature data is independently analyzed, and tasks are selectively directed to regions with lower accumulated thermal wear, creating localized optimization rather than uniform treatment.
Solution Approach 2:
The system changes the parameter of task distribution by using thermal wear data to dynamically adjust which die regions receive which tasks. Instead of fixed task allocation or allocation based solely on immediate temperature, the system modifies task distribution parameters based on cumulative thermal wear history, balancing performance with long-term reliability.
3Reliability
If thermal control logic operates without persistent memory storage, then device complexity is reduced, but thermal wear leveling cannot be implemented
Solution Approach 1:
The system uses copying by maintaining a persistent memory copy of historical temperature data separately from the operational logic. This copied thermal history information is stored in non-volatile memory, allowing the thermal control logic to access accumulated wear data without requiring complex real-time sensing and processing infrastructure during normal operation.
Data Source
AI summary
Exemplary embodiments provide thermal wear spreading among a plurality of thermal die regions in an integrated circuit or among dies by using die region wear-out data that represents a cumulative amount of time each of a number of thermal die regions in one or more dies has spent at a particular temperature level. In one example, die region wear-out data is stored in persistent memory and is accrued over a life of each respective thermal region so that a long term monitoring of temperature levels in the various die regions is used to spread thermal wear among the thermal die regions. In one example, spreading thermal wear is done by controlling task execution such as thread execution among one or more processing cores, dies and/or data access operations for a memory.