Dynamic Swarm Segmentation for Peer Matching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In peer-to-peer computing systems, the inefficiency arises from peers having non-overlapping byte ranges for software updates, leading to ineffective content sharing due to lack of overlapping data, which hampers the effectiveness of the network in distributing updates efficiently.

Innovation Solution

A system utilizing a graph representation of datasets to identify and match clients based on their required byte ranges, allowing for efficient peer matching by grouping clients with similar update needs and traversing the graph to find suitable peers, even if ideal matches are not available within the same group.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If peers are matched based on having any content overlap, then peer-to-peer content sharing can occur, but the matching efficiency is low and many peers become ineffective when byte ranges do not overlap

Engineering Contradiction:
Improvecontent sharing efficiencyVSAvoidpeer matching complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the content into distinct byte ranges and organizes them in a hierarchical graph structure. Each node in the graph represents a specific byte range, allowing the system to break down the complex peer matching problem into smaller, manageable segments. This enables efficient identification of overlapping byte ranges between peers without requiring complex global analysis.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a hierarchical dimension to the peer matching process by organizing byte ranges in a graph structure with multiple levels. Instead of performing flat, two-dimensional peer comparisons, the system traverses the hierarchical graph to find matches, adding a dimensional aspect that simplifies the matching logic and improves efficiency.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Reliability

If the entire dataset is distributed to all peers, then all peers can have complete content, but the data transfer volume and network load increase significantly

Engineering Contradiction:
Improvecontent completenessVSAvoidnetwork bandwidth consumption
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The patent extracts and identifies only the specific byte ranges that each peer needs, rather than distributing the entire dataset. By using the hierarchical graph to determine exact content requirements, the system extracts only the necessary portions of data for transfer, significantly reducing network bandwidth consumption while ensuring each peer receives complete content.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies local quality by tailoring the data distribution to each peer's specific needs. Instead of uniform distribution of all data to all peers, the system determines the local requirements of each peer and provides only those specific byte ranges, optimizing network efficiency while maintaining content completeness for each individual peer.

Inventive Principle:
Principle #3Local quality

3Productivity

If peers trade all their content with each other, then content sharing is maximized, but the time required for content exchange increases due to unnecessary data transfer

Engineering Contradiction:
Improvecontent sharing throughputVSAvoidcontent exchange time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent performs preliminary action by pre-organizing the content structure into a hierarchical graph of byte ranges before peer matching occurs. This pre-structured organization allows the system to quickly identify which byte ranges need to be exchanged, eliminating the need for time-consuming trial-and-error content sharing and enabling peers to exchange only the necessary data immediately.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces dynamics by enabling flexible, adaptive peer matching based on real-time content requirements. The hierarchical graph structure allows the system to dynamically determine the optimal peer pairs and byte range exchanges, adapting to different peer configurations and content needs rather than following a fixed, static distribution pattern.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3298754B1Dynamic swarm segmentation
Publication Date: 2020.03.25 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP3298754B1 patent drawingFigure 1
  • EP3298754B1 patent drawingFigure 2A
  • EP3298754B1 patent drawingFigure 2B

AI summary

Identifying peers to a client for the client to obtain data. A method includes receiving from the client an identification of a dataset and a specification of one or more byte ranges of the dataset. As a result, the method further includes identifying one or more other clients associated with the one or more byte ranges of the dataset to acts as peers to the client. The method further includes providing an indication of the one or more of the other identified clients as peers to the client.