Dynamic Task Assignment in Distributed Data Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In highly distributed data processing systems, maintaining a consistent routing table for task partitioning is challenging, especially during updates, leading to potential incorrect assignments and inefficiencies in resource management.

Innovation Solution

A method that dynamically adjusts task responsibilities among computing components by identifying triggers for partition division, using control units to manage source and target components, and employing a gossip protocol for consistent updates, allowing for efficient redistribution of tasks and data while ensuring continuous processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a routing table is maintained for task partitioning in a distributed system, then task allocation can be managed, but consistency errors occur during updates leading to incorrect assignments

Engineering Contradiction:
Improvetask assignment accuracyVSAvoidrouting table consistency management
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent divides the routing table into multiple independent segments or views that can be updated separately. Instead of maintaining a single global routing table that requires consistent updates across all components, the system segments the routing information into component-specific views that can be independently managed and updated without affecting the entire system's consistency.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Each computing component maintains its own local routing table view with locally optimized consistency properties. Rather than requiring global consistency across all components, each component's routing table is consistent within its local context, allowing updates to propagate independently without requiring system-wide coordination.

Inventive Principle:
Principle #3Local quality

2Adaptability or versatility

If computing components are added or removed from the system, then system scalability is improved, but manual reconfiguration is required which reduces operational efficiency

Engineering Contradiction:
Improvesystem scalabilityVSAvoidmanual reconfiguration effort
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The system automatically detects when computing components are added or removed and self-reconfigures the routing tables without manual intervention. The routing table management system monitors system changes and automatically updates the appropriate routing views to reflect the new system state, eliminating the need for operators to manually reconfigure the system.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent implements preliminary routing table structures that are designed to accommodate dynamic component addition and removal. The routing tables are pre-configured with flexible structures that can automatically adapt to system changes, allowing new components to be integrated seamlessly without requiring manual reconfiguration steps.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If task partitioning is performed with fine granularity, then resource utilization efficiency is improved, but the complexity of maintaining consistent routing information increases

Engineering Contradiction:
Improveresource utilization efficiencyVSAvoidrouting table maintenance complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The routing table is segmented into manageable sections corresponding to different task partitions or component groups. This segmentation allows the system to maintain fine-grained task partitioning for improved resource utilization while reducing the complexity of maintaining routing information by organizing it into discrete, independently manageable segments.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3475819B1Method for automatically and dynamically assigning the responsibility for tasks to the available computing components in a highly distributed data-processing system
Publication Date: 2021.08.11 BULL SA
  • EP3475819B1 patent drawingFigure 1~1A
  • EP3475819B1 patent drawingFigure 1B

AI summary

The invention relates to a method for automatically and dynamically assigning the responsibility for tasks to the available computing components in a highly distributed data-processing system. The invention relates to a method by means of which the assignment of the responsibility for certain tasks (partitions) is adapted in a highly distributed data-processing system.