Distributed Benchmarking System for Client Device Performance Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The diverse range of client devices, exceeding 25,000 types, makes manual determination of performance metrics impractical, and existing methods fail to accurately assess device performance in real-world conditions due to varying usage characteristics and environmental factors.
Innovation Solution
A distributed benchmarking system that distributes benchmark applications to client devices to gather performance metrics, groups them by device aspects like model number, CPU, and GPU type, and uses fuzzy logic to categorize unknown devices based on similarity with known types, reducing the processing burden on individual devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual determination of performance metrics is used, then accuracy can be maintained, but it becomes impractical due to the diverse range of 25,000+ client devices
Solution Approach 1:
The patent uses virtual device profiles that copy and represent groups of physical devices with similar characteristics. Instead of manually testing each of the 25,000+ device types, the system creates virtual representations that capture the essential performance characteristics, allowing metrics to be determined efficiently while maintaining accuracy through the copying approach.
Solution Approach 2:
The system enables client devices to self-report their performance metrics through automated benchmarking applications. Devices automatically execute tests and submit results to the server, eliminating the need for manual determination while scaling to handle diverse device types. This self-service mechanism allows productivity to increase without sacrificing measurement precision.
2Loss of information
If benchmark applications are executed on all client devices, then comprehensive performance data can be gathered, but it overburdens individual devices
Solution Approach 1:
The patent segments the benchmarking process by grouping devices into categories based on shared characteristics (device type, OS version, hardware specifications). Instead of treating all devices uniformly, the system divides them into segments and applies targeted benchmarking strategies to each group, reducing the overall processing burden while maintaining data completeness.
Solution Approach 2:
The system implements selective benchmarking where not all devices execute all benchmark applications. Based on device profiles and categorization, the server determines which subset of benchmarks should run on each device type. This partial action approach ensures comprehensive data gathering across device categories without overburdening individual devices with unnecessary tests.
3Measurement precision
If performance metrics are determined for each individual device type, then precision is improved, but device complexity increases
Solution Approach 1:
The patent merges multiple individual device profiles into consolidated device type categories. By combining devices with similar characteristics into unified groups, the system reduces the number of separate profiles that need to be managed while still maintaining precise metrics for each category. This merging approach decreases system complexity without sacrificing the ability to determine accurate performance metrics for each device type group.
Solution Approach 2:
The virtual device profiles serve multiple functions: they represent groups of physical devices, store performance metrics, guide benchmarking decisions, and enable feature configuration. This multi-functionality reduces the need for separate systems for each function, thereby reducing overall system complexity while maintaining precision in performance determination.
Data Source
AI summary
Systems, devices, media, and methods are presented for determining performance metrics of client devices on a network using benchmark applications. Benchmark applications are distributed to client devices to produce performance metrics for the client devices. Performance metrics of the client devices received from the client devices are grouped to obtain performance metrics for different types of client devices.