Adaptive Memory RAS Policy Switching for Correlated Error Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional RAS policies in computer systems are suboptimal due to spatial and temporal variations in memory errors, leading to issues like silent data corruption and double device errors, which are not effectively addressed by static selection methods.
Innovation Solution
An adaptive RAS platform that dynamically adjusts memory RAS policies at runtime based on correlative prediction, leveraging predictive factors such as temperature, access patterns, and error rates to optimize error correction and detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional RAS policies (SDDC, memory mirroring) are used, then basic error correction is provided, but double device errors under high temperature conditions lead to silent data errors that cannot be corrected
Solution Approach 1:
The patent implements dynamic RAS policy adjustment by continuously monitoring memory error patterns and adapting the RAS scheme in real-time. The system transitions from static conventional RAS policies to dynamic adaptive policies that respond to changing temperature conditions and error characteristics, enabling the system to handle double device errors that occur under high temperature stress
Solution Approach 2:
The system changes operational parameters by monitoring temperature conditions and error rates, then adjusting RAS policy parameters accordingly. When double device errors are detected or predicted, the system modifies correction parameters to apply more aggressive error correction strategies, thereby adapting to the deteriorating memory reliability under high temperature conditions
2Reliability
If adaptive RAS platform with correlative prediction is implemented, then silent data errors are prevented by dynamically adjusting RAS schemes, but system complexity increases
Solution Approach 1:
The patent implements feedback mechanisms by continuously monitoring memory error patterns, temperature conditions, and RAS policy effectiveness. The correlative prediction system uses this feedback to predict future error patterns and proactively adjust RAS policies before silent data errors occur, creating a closed-loop control system that balances reliability improvement with manageable complexity
Solution Approach 2:
The adaptive RAS platform performs self-adjustment by automatically monitoring its own performance and error patterns, then autonomously selecting and applying appropriate RAS policies without external intervention. The system serves itself by maintaining an inventory of RAS schemes and automatically selecting the most appropriate scheme based on current conditions, reducing the need for complex external management
3Reliability
If multiple RAS schemes are maintained and dynamically selected, then error correction effectiveness is improved, but performance overhead increases
Solution Approach 1:
The patent applies partial action by maintaining a selective inventory of RAS schemes rather than applying all possible correction methods continuously. The correlative prediction system identifies when aggressive error correction is actually needed versus when lighter RAS policies suffice, applying excessive correction only when necessary to prevent silent data errors while maintaining performance during normal operation
Data Source
AI summary
An embodiment of an electronic apparatus may comprise one or more substrates and a controller coupled to the one or more substrates, the controller including circuitry to apply a reliability, availability, and serviceability (RAS) policy for access to a memory in accordance with a first RAS scheme, change the applied RAS policy in accordance with a second RAS scheme at runtime, where the second RAS scheme is different from the first RAS scheme, and access the memory in accordance with the applied RAS policy. Other embodiments are disclosed and claimed.