Network Resource Discovery via Concurrent Task Scheduling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current network management systems are inefficient in discovering and processing data from network devices, leading to wastage of computing and human resources due to time-consuming and costly execution of discovery tasks, which also result in poor utilization of computing resources.

Innovation Solution

A network management system utilizing a concurrent discovery model and capacity scheduler model to process network data, determining schedules based on weights for executing discovery tasks, and efficiently managing queues to conserve resources by executing tasks in parallel or sequentially, thereby optimizing resource utilization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If discovery tasks are executed sequentially in traditional network management systems, then task execution is simple, but resource utilization is poor and time consumption increases

Engineering Contradiction:
Improveresource utilization efficiencyVSAvoidtask scheduling complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments discovery tasks into multiple queues (first queue for first discovery tasks, second queue for second discovery tasks) and further divides them into batches. This segmentation enables parallel execution of tasks while maintaining manageable complexity through structured organization, directly resolving the contradiction between improved productivity and scheduling complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary actions by pre-processing network data to identify and categorize discovery tasks before execution. Tasks are organized into queues and batches in advance, allowing the system to execute multiple tasks concurrently without increasing operational complexity during execution, thus improving resource utilization efficiency.

Inventive Principle:
Principle #10Preliminary action

2Loss of time

If all discovery tasks are executed concurrently without scheduling, then discovery speed increases, but computing resources are wasted

Engineering Contradiction:
Improvediscovery timeVSAvoidcomputing resource waste
Core Design Contradiction:
Loss of timeVSLoss of energy

Solution Approach 1:

The patent implements dynamic scheduling by adjusting task execution based on real-time conditions. The capacity scheduler dynamically determines which batches to execute and when, adapting to varying network conditions and resource availability. This dynamic approach reduces discovery time while preventing computing resource waste by only executing tasks when resources are available and needed.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes execution parameters by varying the concurrency level and batch size based on system capacity. The capacity scheduler modifies parameters such as the number of simultaneous tasks and resource allocation dynamically, optimizing the balance between discovery speed and resource utilization, thereby reducing both time loss and energy waste.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If discovery tasks are executed in batches with capacity scheduling, then resource utilization improves, but task execution time increases

Engineering Contradiction:
Improvecomputing resource utilizationVSAvoidtask execution duration
Core Design Contradiction:
ProductivityVSDuration of action of moving object

Solution Approach 1:

The patent ensures continuity of useful action by maintaining multiple queues of discovery tasks that can be executed continuously as resources become available. The capacity scheduler continuously monitors system capacity and launches batches of tasks without idle periods, ensuring that computing resources are continuously utilized productively while minimizing total execution duration through overlapping task execution.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentEP3985506B1Utilizing models for concurrently discovering network resources of a network
Publication Date: 2024.05.29 JUNIPER NETWORKS INC
  • EP3985506B1 patent drawingFigure 1A
  • EP3985506B1 patent drawingFigure 1B
  • EP3985506B1 patent drawingFigure 1C

AI summary

A device may receive network data associated with network devices, and may process the network data, with a first model, to determine a first queue identifying first discovery tasks to execute for the network devices and a second queue identifying second discovery tasks to execute for the network devices. The device may process the first queue and the second queue, with a second model, to determine a schedule for executing the first and second discovery tasks, and may execute the first and second discovery tasks based on the schedule. The device may calculate progress of the executions of the first discovery tasks and the second discovery tasks, and may estimate time intervals associated with completions of the executions of the first discovery tasks and the second discovery tasks. The device may provide, to a user device, information identifying the progress and the time intervals.