Adaptive DRAM Power Management via Usage-Based Weights
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing DRAM power saving technologies, such as PASR, PAAR, and DPD, do not result in optimal power savings as they often require aggressive shutdowns that can degrade performance and battery life, due to static settings that do not adapt to user behavior.
Innovation Solution
An adaptive memory controller that collects usage data and dynamically adjusts PASR, PAAR, and DPD settings based on weights associated with different usage patterns, optimizing power management by controlling memory refresh operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conservative refresh settings are used to prevent data loss, then data reliability is improved, but energy consumption increases
Solution Approach 1:
The patent implements dynamic refresh interval adjustment based on temperature conditions. The memory controller monitors temperature and adaptively changes the refresh interval - using shorter intervals at high temperatures to prevent data loss, and longer intervals at low temperatures to reduce energy consumption. This dynamic adaptation resolves the contradiction between maintaining data reliability and minimizing energy usage.
Solution Approach 2:
The system changes the refresh interval parameter based on temperature conditions. By monitoring temperature and adjusting the refresh interval parameter accordingly, the system optimizes the balance between data reliability (preventing data loss) and energy consumption (reducing unnecessary refreshes).
2Use of energy by moving object
If PASR, PAAR, or DPD are engaged frequently to save power, then energy savings increase, but performance degrades due to excessive CPU subtasks
Solution Approach 1:
The patent employs feedback mechanisms where the system monitors temperature conditions and CPU activity levels before engaging power-saving modes. By continuously monitoring system state and adjusting power management decisions based on this feedback, the system avoids excessive engagement of PASR/PAAR/DPD that would degrade performance while still achieving meaningful energy savings when conditions are appropriate.
Solution Approach 2:
The system dynamically adjusts the frequency and aggressiveness of power-saving mode engagement based on real-time temperature and workload conditions. Rather than frequently engaging PASR/PAAR/DPD regardless of conditions, the system adapts its power management strategy to current system state, resolving the contradiction between energy savings and performance.
3Device complexity
If static PASR, PAAR, and DPD settings are used, then device complexity is reduced, but power savings are suboptimal due to inability to adapt to user behavior
Solution Approach 1:
The patent implements dynamic power management that adapts to user behavior patterns. The system monitors usage patterns and temperature conditions, then dynamically adjusts refresh strategies and power-saving mode engagement. This dynamic approach achieves optimal power savings without requiring complex user configuration, as the system automatically adapts to observed usage patterns.
Solution Approach 2:
The system performs self-adjustment of power management parameters based on monitored usage patterns and temperature conditions. Rather than requiring user configuration or complex external control, the memory controller autonomously optimizes refresh strategies and power-saving mode engagement based on observed system state and usage patterns.
4Reliability
If refresh interval is shortened to prevent data loss at high temperatures, then data reliability is improved, but unnecessary energy is expended
Solution Approach 1:
The system dynamically changes the refresh interval parameter based on monitored temperature conditions. When temperature is high, the refresh interval is shortened to prevent data loss. When temperature is low, the refresh interval is extended to reduce unnecessary energy expenditure. This parameter adaptation based on environmental conditions resolves the contradiction between data reliability and energy efficiency.
Solution Approach 2:
The patent implements a dynamic refresh strategy where the refresh interval is continuously adjusted based on temperature monitoring. This dynamic adaptation ensures data reliability when needed (high temperatures) while minimizing energy waste when conditions permit (low temperatures).
Data Source
AI summary
Various additional and alternative aspects are described herein. In some aspects, the present disclosure provides a method of controlling a memory of a computing device by an adaptive memory controller. The method includes collecting usage data from the computing device over a first bin, wherein the first bin is associated with a first weight, wherein the first weight is indicative of one or more of a first partial array self-refresh (PASR) setting a first partial array auto refresh (PAAR) setting and a first deep power down (DPD) setting. The method further includes associating the collected data with a second weight, adapting the first bin based on the second weight, wherein the second weight is indicative of one or more of a second PASR, PAAR, and DPD setting. The method further includes controlling the memory during the next first bin based on the second weight.