Task Scheduling System for Heterogeneous Resource Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current systems face challenges in efficiently managing task concurrency on heterogeneous computer platforms, particularly in optimizing the distribution and scheduling of concurrent tasks across diverse computing resources, leading to suboptimal utilization of available resources.

Innovation Solution

A task scheduling system that automates the distribution of data and binding of tasks to computing resources, utilizing a discoverer component to identify dependencies and a tuner component to optimize task ordering and resource allocation, allowing for dynamic scheduling and efficient use of multiple processor cores and memory devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual task distribution and scheduling is used on heterogeneous platforms, then developers have control over task allocation, but resource utilization becomes suboptimal and management complexity increases

Engineering Contradiction:

Improveresource utilizationVSAvoidmanagement complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system enables automated self-scheduling of concurrent tasks across heterogeneous computing resources. The task scheduling system automatically discovers dependencies, allocates tasks to appropriate resources, and optimizes execution without requiring manual intervention, thereby improving resource utilization while reducing management complexity

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system dynamically adjusts scheduling parameters and task distribution strategies based on real-time platform conditions and resource availability. By changing scheduling parameters adaptively, the system optimizes resource utilization across heterogeneous platforms without increasing management burden

Inventive Principle:

Principle #35Parameter changes

2Productivity

If concurrent tasks are executed without automated scheduling, then task execution simplicity is maintained, but resource distribution efficiency decreases

Engineering Contradiction:

Improvetask distribution efficiencyVSAvoidscheduling simplicity

Core Design Contradiction:

ProductivityVSEase of operation

Solution Approach 1:

The system performs preliminary automated analysis of task dependencies and resource requirements before execution. By pre-processing task graphs and identifying optimal scheduling strategies in advance, the system improves task distribution efficiency while maintaining operational simplicity through automated decision-making

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The task scheduling system acts as an intermediary layer between task submission and execution. It automatically manages the complexity of concurrent task coordination, dependency resolution, and resource allocation, thereby improving distribution efficiency while shielding users from scheduling complexity

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If heterogeneous computing resources are used, then computing power and capability are increased, but task mapping and resource allocation complexity increases

Engineering Contradiction:

Improvecomputing resource capabilityVSAvoidtask mapping complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system applies specialized scheduling strategies tailored to different types of computing resources (CPU, GPU, FPGA, etc.). By optimizing task mapping for each resource type's specific capabilities and characteristics, the system leverages heterogeneous computing power while managing mapping complexity through resource-specific optimization rules

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The task scheduling system provides a universal interface and automated mapping mechanism that works across diverse heterogeneous resources. It consolidates complex resource-specific mapping logic into a unified scheduling framework, enabling versatile resource utilization without exposing mapping complexity to users

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS10474510B2Declarative properties for data collections

Publication Date: 2019.11.12 INTEL CORP

AI summary

A system according one exemplary embodiment may receive instructions regarding a callee function that include a description of properties associated with arguments for the callee function, create a persistent representation of the callee function based on the received description of the callee function, receive instructions from a source describing a set of properties for a data collection, create a persistent representation of the data collection based on the set of properties for the data collection, receive an updated property for the data collection, and in response to receiving the updated property for the data collection: update the persistent representation of the callee function by associating the persistent representation of the callee function with the updated property of the data collection and update the persistent representation of the data collection by associating the persistent representation the data collection with the updated property of the data collection.