Predictive Analytics Model for Edge Device Query Enrichment

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Edge devices face challenges in managing and analyzing large volumes of data due to limited storage and computational capabilities, and frequent requests for enrichment data are hindered by poor network connections, leading to inefficiencies and server overload.

Innovation Solution

A system that iteratively updates local instructions at edge devices using global data aggregated by a central server, enabling efficient data processing and query evaluation, and implements intelligent throttling to reduce data transmission, allowing edge devices to determine necessary enrichment data and transmit it efficiently.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If edge devices frequently request enrichment data from the server, then query evaluation accuracy is improved, but network load increases and server becomes overloaded

Engineering Contradiction:

Improvequery evaluation accuracyVSAvoidserver capacity

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The server proactively pushes enrichment data to edge devices before queries are executed. The server maintains a cache of enrichment data and identifies which devices need updates based on their last query times and data expiration policies, delivering relevant data in advance to avoid frequent requests during network poor conditions.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements a feedback mechanism where the server monitors query execution results and adjusts its data pushing strategy accordingly. Successful queries and their performance metrics are fed back to the server, which uses this information to optimize future data delivery decisions and determine which edge devices need enrichment data updates.

Inventive Principle:

Principle #23Feedback

2Quantity of substance

If edge devices transmit all ingested data to the server, then data analysis completeness is improved, but network bandwidth consumption increases

Engineering Contradiction:

Improvedata analysis completenessVSAvoidnetwork bandwidth consumption

Core Design Contradiction:

Quantity of substanceVSLoss of energy

Solution Approach 1:

The system extracts and transmits only the essential data elements needed for effective analysis rather than all ingested data. The server identifies critical data patterns and queries, then pushes only the relevant enrichment data to edge devices, reducing unnecessary data transmission while maintaining analysis completeness.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system dynamically adjusts data transmission parameters based on network conditions and query requirements. The server modifies data pushing frequency, volume, and timing based on real-time network status and edge device needs, optimizing the balance between data completeness and bandwidth consumption.

Inventive Principle:

Principle #35Parameter changes

3Adaptability or versatility

If edge devices perform comprehensive data processing locally, then data processing flexibility is improved, but device computational capability is exceeded

Engineering Contradiction:

Improvedata processing flexibilityVSAvoiddevice computational capability

Core Design Contradiction:

Adaptability or versatilityVSPower

Solution Approach 1:

The system segments data processing tasks between edge devices and the server. Edge devices perform lightweight local processing and query execution, while the server handles computationally intensive data enrichment, aggregation, and complex analysis. This division allows comprehensive processing capability without overloading individual edge devices.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The server acts as an intermediary that provides computational support to edge devices. Instead of requiring edge devices to perform all processing independently, the server mediates by providing pre-processed data, execution plans, and computational results, enabling flexible processing within edge device capability limits.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11843505B1System and method of generation of a predictive analytics model and performance of centralized analytics therewith

Publication Date: 2023.12.12 CISCO TECHNOLOGY INC

AI summary

A computerized method is disclosed that includes operations of receiving one or more records, wherein each of the one or more records indicates a successful search query evaluation by at least one of a plurality edge devices, building a predictive analytics model based on the one or more records, wherein the predicative analytics model is configured to perform operations configured to predict enrichment data that is to be needed by one or more edge devices in the future during evaluation of a future search query, performing predictive analytics using the predictive analytics model to determine predictive enrichment data, and transmitting a first response packet to a first edge device, wherein the first response packet includes the predictive enrichment data. The records may include one or more of a data stream identifier, a search query, enrichment data that was required at a time the search query was evaluated.