IoT Platform Performance Prediction Using Queuing Network Models

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current IoT platforms lack effective performance prediction methods to handle dynamic and growing data sets from smart devices, leading to uncertainties in performance, reliability, and scalability, especially when dealing with large volumes of user requests and sensor observations.

Innovation Solution

A method and system for predicting IoT application performance by obtaining user requests and sensor observations, identifying API flows, measuring resource utilization, and using queuing network models to compute service demands and predict performance under varying workload conditions, incorporating performance testing and log analysis to derive inter-arrival distributions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional performance modeling techniques are applied to IoT applications, then the model can handle fixed size data sets, but it cannot address larger volume of dynamic data sets from smart devices

Engineering Contradiction:

Improveadaptability to dynamic data setsVSAvoidperformance prediction accuracy

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent applies dynamics by transitioning from static performance modeling to dynamic performance modeling that adapts to changing workload characteristics. The system continuously monitors resource utilization and adjusts performance models in real-time to accommodate dynamic data sets from smart devices, making the model flexible rather than fixed.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent changes key parameters including workload characteristics, resource utilization metrics, and performance thresholds. By dynamically adjusting these parameters based on actual system state, the model can adapt to varying data volumes while maintaining prediction accuracy through continuous parameter optimization.

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If multiple software components and technologies are integrated in IoT device gateway tier, then functionality is enhanced, but complexity in building performance models increases

Engineering Contradiction:

Improvefunctional capabilityVSAvoidperformance model complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the complex IoT system into distinct modular components including device gateway tier, platform tier, and application tier. Each component has its own performance model that can be independently developed and maintained. This segmentation reduces overall model complexity while preserving full functionality across all software components.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces performance modeling intermediaries that act as abstraction layers between multiple software components and the overall performance model. These intermediaries standardize interfaces and communication protocols, making it easier to integrate diverse technologies without proportionally increasing model complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Measurement precision

If performance testing is conducted with large volume of workload characteristics, then prediction accuracy improves, but testing time and computational resources increase

Engineering Contradiction:

Improveperformance prediction accuracyVSAvoidtesting time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary performance testing with representative workload characteristics to establish baseline performance models before full-scale deployment. By conducting preliminary tests with carefully selected workload scenarios, the system achieves sufficient prediction accuracy without requiring exhaustive testing of all possible workload volumes, thus reducing overall testing time.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent applies partial action by conducting performance tests with a strategically selected subset of workload characteristics rather than all possible volumes. The testing focuses on critical thresholds and representative scenarios that provide sufficient prediction accuracy for decision-making, avoiding the time cost of exhaustive testing while maintaining adequate precision.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentEP3296876B1Systems and methods for predicting performance of applications on an internet of things (IOT) platform

Publication Date: 2019.08.21 TATA CONSULTANCY SERVICES LTD

Figure 1
Figure 2
Figure 3

AI summary

Performance prediction systems and method of an Internet of Things (IoT) platform and applications includes obtaining input(s) comprising one of (i) user requests and (ii) sensor observations from sensor(s); invoking Application Programming Interface (APIs) of the platform based on input(s); identifying open flow (OF) and closed flow (CF) requests of system(s) connected to the platform; identifying workload characteristics of the OF and CF requests to obtain segregated OF and segregated CF requests, and a combination of open and closed flow requests; executing performance tests with the APIs based on the workload characteristics; measuring resource utilization of the system(s) and computing service demands of resource(s) from measured utilization, and user requests processed by the platform per unit time; executing the performance tests with the invoked APIs based on volume of workload characteristics pertaining to the application(s); and predicting, using queuing network, performance of the application(s) for the volume of workload characteristics.