Network Nodes Recovering Missing Data Units From Peer Buffers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Nodes in a network may fail to capture and store real-time broadcast data feeds, resulting in missing data that cannot be recovered since the data source does not re-transmit missing data units, leading to data loss for clients.

Innovation Solution

Implementing a method where nodes identify data lacunas by determining the extent of missing data units and request corresponding data from other nodes that have received the complete data feed, allowing for self-healing by filling in the gaps without requiring re-transmission from the source.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a node captures and stores real-time broadcast data feed, then data availability is improved, but data loss occurs when the node fails to capture data due to network issues or buffer capacity limits

Engineering Contradiction:
Improvedata availabilityVSAvoiddata loss
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

Other nodes in the network proactively capture and store data units from the broadcast data feed before the first node needs them. When the first node detects missing data units, these data units are already available at other nodes, allowing immediate recovery without requiring re-transmission from the broadcast source.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

Other nodes act as intermediaries by holding copies of data units and transferring them to the first node when needed. This intermediary mechanism allows data recovery without direct communication with the original broadcast source, resolving the contradiction between data availability and data loss.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If the data source only broadcasts data in real-time without re-transmission, then transmission efficiency is improved, but missing data cannot be recovered

Engineering Contradiction:
Improvetransmission efficiencyVSAvoiddata recovery
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

Multiple nodes create copies of the broadcast data feed by capturing and storing data units independently. When one node loses data, it can retrieve copies from other nodes that successfully captured the data, maintaining data recovery capability without requiring the source to re-transmit.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

Nodes pre-capture and store data units from the broadcast feed in advance, creating a distributed backup system. This preliminary action ensures that data copies are available before any node experiences data loss, enabling recovery without source re-transmission.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If a node requests missing data from other nodes, then data completeness is improved, but network communication overhead increases

Engineering Contradiction:
Improvedata completenessVSAvoidcommunication overhead
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The first node monitors the sequence numbers of received data units and generates feedback requests for missing data units. This feedback mechanism enables automatic detection and recovery of missing data by requesting only the specific units that are absent, minimizing unnecessary communication.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

Other nodes maintain buffers of captured data units in advance, so when the first node requests missing data, the requested units are immediately available for transfer without requiring additional processing or retrieval operations, reducing communication overhead.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3031172B1Managing data feeds
Publication Date: 2017.06.28 AB INITIO TECHNOLOGY LLC
  • EP3031172B1 patent drawingFigure 1~2
  • EP3031172B1 patent drawingFigure 3
  • EP3031172B1 patent drawingFigure 4

AI summary

Managing data units broadcast from a data feed (14), without re-transmission by a source of the data feed, includes: at a first node in a network (12), receiving at least a portion of a data feed including a plurality of data units; at a second node in the network, receiving at least a portion of the data feed; identifying an interruption in receiving the data feed at the first node; determining an extent of a data lacuna (504) extending between a last data unit received by the first node prior to the interruption and a first data unit received by the first node after the interruption; and sending (506) a request from the first node for results (510) saved by the second node, the results saved by the second node corresponding to the data lacuna.