Estimating Application Performance via Network Traffic Ratios

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Hardware vendors face challenges in designing processors and servers for improved performance and scalability, as they lack a priori knowledge of specific application software behaviors, making it difficult to implement effective real-time performance monitoring and optimization without cooperation from software vendors.

Innovation Solution

A system that estimates application performance variations by collecting and analyzing network traffic statistics, calculating ratios of network traffic data across sampling intervals, and using this data to estimate application execution performance without requiring detailed knowledge of the application's operational details.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If hardware vendors implement real-time application performance monitoring, then performance optimization capability is improved, but system complexity and implementation difficulty increase due to lack of application knowledge

Engineering Contradiction:

Improveperformance optimization capabilityVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces network traffic statistics as an intermediary that indirectly reflects application performance. Instead of directly monitoring application execution details (which requires application knowledge), the system monitors network traffic patterns that correlate with performance, thereby simplifying the monitoring system while maintaining optimization capability

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces complex application-level performance monitoring mechanisms with simpler network traffic analysis. By substituting direct application performance measurement (which requires deep application knowledge) with network traffic statistic analysis, the system reduces complexity while preserving the ability to detect performance variations

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Adaptability or versatility

If hardware vendors use general-purpose performance monitoring without application knowledge, then solution versatility is improved, but measurement precision deteriorates

Engineering Contradiction:

Improvesolution versatilityVSAvoidperformance measurement precision

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent establishes a feedback mechanism where network traffic statistics continuously reflect application performance state. By monitoring changes in network traffic patterns over time and using this feedback to infer performance variations, the system achieves both general applicability and sufficient measurement precision for dynamic optimization

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent monitors changes in network traffic parameters (such as bandwidth utilization, packet rates) as proxies for application performance changes. By tracking parameter variations rather than absolute values, and using ratio calculations between sampling intervals, the system achieves precision in detecting performance variations without requiring absolute performance measurements

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If hardware vendors require cooperation from software vendors for performance monitoring, then measurement precision is improved, but ease of operation deteriorates

Engineering Contradiction:

Improveperformance measurement precisionVSAvoidimplementation ease

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The patent enables the hardware monitoring system to independently gather necessary performance information through network traffic statistics without requiring software vendor cooperation or installation of application-specific monitoring tools. The system serves itself by using readily available network interface data to infer application performance

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS9954744B2Estimation of application performance variation without a priori knowledge of the application

Publication Date: 2018.04.24 INTEL CORP

AI summary

Generally, this disclosure provides systems, devices, methods and computer readable media for estimation of application execution performance variations on a processor, without a priori knowledge of the application. A system may include network traffic data collection circuitry configured to sample a first network traffic statistic, from a network interface circuit associated with the processor, at a first sampling time interval during the application execution. The network traffic data collection circuitry may also be configured to sample a second network traffic statistic from the network interface circuit at a second sampling time interval during the application execution. The system may further include performance analysis circuitry configured to calculate a ratio of the first network traffic statistic to the second network traffic statistic and to estimate the application execution performance variation from the first sampling time interval to the second sampling time interval, wherein the estimation is proportional to the calculated ratio.