Requirements-Based Performance Monitor for Virtual Machine I/O
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In virtualized computing systems, existing performance monitoring tools fail to effectively account for application requirements, leading to difficulties in determining whether I/O needs are met, and often blame either the system or the application for performance issues, making it challenging to maintain acceptable performance levels.
Innovation Solution
A requirements-based performance monitoring (RBPM) process that monitors device operating parameters and compares them to application requirements, providing a user-friendly interface to gauge if I/O needs are met, using metrics like IOPS, throughput, and latency, and generating performance scores based on a 1-100% scale to indicate system health.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional performance monitoring tools are used in virtualized systems, then comprehensive data collection is achieved, but the ability to determine whether application I/O needs are met deteriorates
Solution Approach 1:
The patent introduces a performance monitor component that acts as an intermediary between the hypervisor and the application. This mediator translates raw performance data into application-specific performance indicators, comparing actual performance against required performance thresholds to determine whether I/O needs are met, thereby resolving the contradiction between comprehensive data collection and accurate performance assessment.
Solution Approach 2:
The patent transforms raw performance parameters (IOPS, throughput, latency) into application-specific performance metrics by changing the parameter representation. It introduces performance thresholds and comparison metrics that are tailored to specific application requirements, enabling accurate assessment of whether I/O needs are met without increasing system complexity.
2Quantity of substance
If multiple performance indicators are collected from hypervisor and guest OS, then data completeness is improved, but ease of operation deteriorates
Solution Approach 1:
The patent extracts only the relevant performance indicators needed for application-specific assessment from the comprehensive set of available data. It separates essential metrics (IOPS, throughput, latency) from extraneous information, and further extracts performance comparisons against thresholds to provide actionable insights, thereby improving ease of operation while maintaining data completeness.
Solution Approach 2:
The patent segments performance monitoring into distinct layers: data collection from hypervisor and guest OS, processing through the performance monitor, and presentation of application-specific performance indicators. This segmentation organizes the complex data flow into manageable segments, making the system easier to operate while preserving comprehensive data collection.
3Measurement precision
If performance monitoring accounts for application requirements, then diagnostic accuracy is improved, but device complexity increases
Solution Approach 1:
The patent implements preliminary action by pre-defining performance thresholds and requirements for different application types before monitoring begins. The performance monitor is pre-configured with expected performance levels, allowing it to directly compare actual performance against these pre-established criteria, thereby improving diagnostic accuracy without adding complex real-time analysis logic.
Data Source
AI summary
Embodiments are directed to a requirements-based performance monitor (RBPM) that presents users, through a command line interface or graphical user interface, with a single number in the range of 1-100% and/or a color-coded indicator that allows users to readily tell if the I/O needs of their primary applications in a virtualized computing system are being sufficiently satisfied. The RBPM takes into account both device latency, throughput, IOPS, and slow I/O measurements and primary application requirements. The process uses detailed device latency tables for each device to allow a user to find the particular device and time that performance degradation occurred.