Distributed Computing Profiles for Job Scheduling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing technical computing applications are limited by their single-workstation design, which restricts the complexity and speed of problem-solving due to processing capacity and time constraints, and transitioning to distributed computing environments is costly and complex.

Innovation Solution

A system and method that allow users to create and utilize profiles, scheduler components, and project components to perform jobs in a distributed computing environment without modifying programming code, enabling execution across multiple devices and facilitating the sharing of these components between devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If technical computing is performed on a single workstation, then the application can be simple and easy to operate, but the processing capacity and problem complexity are limited

Engineering Contradiction:
Improveease of operationVSAvoidprocessing capacity
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The system segments the computing environment into distinct components: profile components (defining computing environment characteristics), scheduler components (managing job distribution), and project components (organizing computational tasks). This segmentation allows the computing workload to be divided across multiple workstations while maintaining a unified interface for users, thus increasing processing capacity without complicating the user experience.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces profile components as intermediaries that act as templates or blueprints for distributed computing environments. These profile components abstract the complexity of distributed system configuration, allowing users to define computing environment characteristics once and reuse them across multiple workstations. The scheduler component serves as another intermediary that automatically distributes jobs between the user and available computing resources based on the profile definitions.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If distributed computing is implemented to increase processing capacity, then productivity improves, but device complexity and implementation cost increase

Engineering Contradiction:
Improveprocessing capacityVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The profile components are designed to be universal templates that can define computing environment characteristics applicable across multiple different workstations and computing resources. A single profile component can be reused to configure identical distributed computing environments on different hardware platforms, reducing the complexity of implementing distributed computing across multiple systems. The scheduler component also provides universal functionality by handling job distribution, monitoring, and result collection across diverse computing resources through a standardized interface.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Productivity

If new technical computing applications are developed for distributed computing, then processing capacity increases, but development cost and time increase

Engineering Contradiction:
Improveprocessing capacityVSAvoiddevelopment cost
Core Design Contradiction:
ProductivityVSEase of manufacture

Solution Approach 1:

The system performs preliminary action by pre-defining computing environment characteristics in profile components before actual distributed computing tasks are executed. These profile components contain all necessary configuration information, scheduler settings, and environment parameters that would otherwise need to be configured for each new application. This preliminary preparation allows existing technical computing applications to be deployed in distributed environments without requiring redevelopment, as the profile components provide the necessary distributed computing infrastructure configuration in advance.

Inventive Principle:
Principle #10Preliminary action

4Speed

If computational tasks are distributed across multiple workstations, then processing speed increases, but system coordination and management become more complex

Engineering Contradiction:
Improveprocessing speedVSAvoidsystem coordination
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The scheduler component implements feedback mechanisms that continuously monitor the state of distributed computing resources and adjust job distribution accordingly. The system collects feedback information about workstation availability, computational load, and task progress, then uses this feedback to dynamically allocate computing tasks across the distributed network. This automated feedback loop simplifies system coordination by replacing complex manual management with intelligent, adaptive resource allocation that responds to real-time system conditions.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP2769302B1Use of dynamic profiles for creating and using a distributed computing environment
Publication Date: 2018.08.29 MATHWORKS INC
  • EP2769302B1 patent drawingFigure 1
  • EP2769302B1 patent drawingFigure 2
  • EP2769302B1 patent drawingFigure 3A~3B

AI summary

A system is configured to create a profile that allows a job to be performed within a distributed computing environment comprising one or more computing devices; provide an interface that allows the profile to be specified; obtain a scheduler component via the interface, where the scheduler component identifies a scheduler to be used to perform the job and an application that at least one of the computing devices, that is specified by the scheduler, is to use when performing the job; obtain, via the interface, a project component that identifies data on which the job is to be based; and create the profile based on the scheduler component and the project component.