Adaptive Thread Scheduling Using Leakage Current Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In asymmetric multiprocessor systems, managing leakage current and dynamic power to achieve high performance at low power expenditure is challenging due to variations in process characteristics, voltage, and temperature, which affect task assignment and energy consumption.
Innovation Solution
An adaptive thread scheduling method that utilizes leakage current information to schedule threads on multiple cores, optimizing frequency and voltage settings to minimize energy usage, with sensors providing real-time data for informed decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple processors operate at high frequency to meet performance requirements, then productivity is improved, but use of energy increases due to increased leakage current
Solution Approach 1:
The patent implements dynamic thread scheduling that adapts to real-time leakage current measurements from multiple processors. The scheduler dynamically assigns threads to processors based on current leakage conditions, allowing the system to optimize the balance between processing performance and energy consumption by redistributing workloads according to instantaneous processor states
Solution Approach 2:
The system changes operational parameters by adjusting thread assignment decisions based on measured leakage current values. By monitoring leakage current as a key parameter and using it to modify scheduling decisions, the system optimizes the trade-off between maintaining high processing throughput and reducing overall energy consumption
2Productivity
If threshold voltage is reduced to increase circuit density and clock frequency, then productivity is improved, but leakage current increases
Solution Approach 1:
The patent implements a feedback mechanism where leakage current is measured from each processor and this information is fed back to the thread scheduler. The scheduler uses this feedback to adjust thread assignments, compensating for the increased leakage current effect by dynamically balancing workloads based on actual leakage conditions
Solution Approach 2:
The system replaces static voltage threshold settings with a dynamic scheduling approach that accounts for leakage variations. Instead of fixing the threshold voltage to maximize frequency, the system substitutes a mechanical/electrical constraint with a software-based scheduling mechanism that adapts to leakage conditions
3Device complexity
If threads are scheduled without considering leakage current variations, then device complexity is reduced, but use of energy increases
Solution Approach 1:
The thread scheduler is enhanced to serve multiple functions: it performs conventional thread scheduling while simultaneously incorporating leakage current monitoring and adaptive workload distribution. This multi-functional approach allows the system to reduce energy consumption without requiring entirely separate mechanisms for leakage management
Data Source
AI summary
Techniques for adaptive thread scheduling on a plurality of cores for reducing system energy are described. In one embodiment, a thread scheduler receives leakage current information associated with the plurality of cores. The leakage current information is employed to schedule a thread on one of the plurality of cores to reduce system energy usage. On chip calibration of the sensors is also described.