Processor Energy Dissipation Estimation for Thermal-Aware Task Scheduling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Data centers face significant energy inefficiencies due to the rapid growth of energy dissipation in servers and cooling systems, with traditional solutions being primarily reactive and failing to optimize energy use across computing and cooling resources effectively.

Innovation Solution

A holistic approach treating energy as a first-class resource, using fast thermal models and global schedulers to allocate energy budgets to servers proactively, combined with modified operating system kernels to manage energy consumption within budgeted limits, and dynamic cooling strategies tailored to specific thermal conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If servers are overprovisioned with high capacity, then peak performance is improved, but energy efficiency deteriorates due to operation at low utilization levels

Engineering Contradiction:
Improvepeak performanceVSAvoidenergy efficiency
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent implements dynamic workload scheduling that actively monitors and adjusts server utilization levels in real-time, transitioning servers between active and standby states based on current workload demands, thereby optimizing energy efficiency while maintaining peak performance capability

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system performs preliminary thermal analysis and workload prediction to proactively schedule tasks before thermal conditions deteriorate, preventing thermal emergencies before they occur and allowing servers to operate at optimal energy efficiency points

Inventive Principle:
Principle #10Preliminary action

2Productivity

If more servers are packed into data centers, then capacity is improved, but energy dissipation and cooling requirements worsen

Engineering Contradiction:
Improvedata center capacityVSAvoidenergy dissipation
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

The patent implements rack-level thermal management that treats each rack as an independent thermal zone with localized monitoring and control, allowing differential cooling strategies and workload scheduling per rack to reduce overall energy dissipation while maintaining high capacity

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system employs continuous thermal monitoring and feedback loops that sense temperature conditions and automatically adjust workload scheduling and cooling resource allocation, creating a closed-loop control system that optimizes energy efficiency at scale

Inventive Principle:
Principle #23Feedback

3Ease of operation

If traditional reactive cooling solutions are used, then response to thermal conditions is simple, but energy optimization across computing and cooling resources deteriorates

Engineering Contradiction:
Improvecooling control simplicityVSAvoidoverall energy efficiency
Core Design Contradiction:
Ease of operationVSLoss of energy

Solution Approach 1:

The patent merges computing workload management with thermal and cooling resource management into a unified scheduling system, where task allocation decisions simultaneously optimize computational efficiency and thermal conditions, achieving overall energy optimization while maintaining operational simplicity

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS12135598B2Apparatus and method for efficient estimation of the energy dissipation of processor based systems
Publication Date: 2024.11.05 THE RES FOUNDATION FOR THE STATE UNIV OF NEW YORK
  • US12135598B2 patent drawing
  • US12135598B2 patent drawing
  • US12135598B2 patent drawing

AI summary

A system and method of scheduling tasks, comprising receiving activity and performance data from registers or storage locations maintained by hardware and an operating system; storing calibration coefficients associated with the activity and performance data; computing an energy dissipation rate based on at least the activity and performance data; and scheduling tasks under the operating system based on the computed energy dissipation rate.