Programmable Logic Thermal-Aware Resource Allocation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Programmable logic devices (PLDs) face inefficiencies and potential overheating due to generic thermal management techniques, which do not account for specific user configurations and thermal profiles, leading to suboptimal performance and operation.

Innovation Solution

Implementing thermal-aware design generation and dynamic resource management within the design software to balance thermal, power, and timing constraints, using techniques such as power gating, clock gating, and custom routing to optimize thermal profiles for specific configurations and workloads.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If generic thermal management techniques are used for PLDs, then thermal control is simplified, but thermal management effectiveness deteriorates due to inability to account for specific user configurations and thermal profiles

Engineering Contradiction:
Improvethermal management complexityVSAvoidthermal management effectiveness
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent implements thermal management by dividing the PLD into multiple regions or zones, each with its own thermal characteristics and configuration-specific thermal profiles. This allows different parts of the device to be managed according to their specific thermal needs rather than applying a uniform generic approach, thereby improving thermal management effectiveness while maintaining reasonable complexity.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent employs dynamic thermal management that adapts to specific user configurations and workloads. The thermal management system dynamically adjusts parameters such as resource allocation, power gating strategies, and clock gating based on real-time thermal conditions and configuration specifics, transforming static generic thermal control into a dynamic responsive system that improves effectiveness without excessive complexity.

Inventive Principle:
Principle #15Dynamics

2Reliability

If dynamic resource management is implemented to optimize thermal profiles, then thermal control effectiveness is improved, but device complexity increases

Engineering Contradiction:
Improvethermal control effectivenessVSAvoidresource management complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements self-service thermal management where the PLD automatically monitors its own thermal conditions and adjusts resource allocation accordingly. The device uses built-in thermal sensors and configuration data to autonomously make decisions about power gating, clock gating, and resource distribution without requiring complex external control systems, thereby improving thermal effectiveness while limiting the increase in overall device complexity.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent employs feedback mechanisms where thermal management decisions are based on real-time thermal condition monitoring and configuration data. The system continuously monitors thermal states, compares them against thresholds and profiles specific to the user configuration, and adjusts resource allocation dynamically. This feedback-driven approach improves thermal control effectiveness while keeping complexity manageable through rule-based decision making rather than complex optimization algorithms.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS20240126228A1Thermal-Aware Programmable Logic Device-Based Programming
Publication Date: 2024.04.18 ALTERA CORP
  • US20240126228A1 patent drawing
  • US20240126228A1 patent drawing
  • US20240126228A1 patent drawing

AI summary

Systems or methods of the present disclosure may provide for implementing design software that is used to design a configuration for a programmable fabric of a programmable logic device. Implementing the design software includes receiving, at a processor, design configuration details for the configuration. Implementing the design software also includes receiving, at the processor, a plurality of constraints including a thermal constraint for the configuration. Moreover, implementing the design software comprises performing thermal aware resource selection based at least in part on the thermal constraint. Furthermore, implementing the design software includes causing the programmable logic device to be operated to stay within the thermal constraint.