VM Instance Optimization Service for Distributed Workloads

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Service providers face challenges in optimizing the selection and configuration of virtual machine (VM) instance types to support diverse workloads efficiently, leading to underutilization or overutilization of computing resources, especially for new users who lack expertise in computing resources.

Innovation Solution

An optimization service within a service provider network that uses workload categorization, resource utilization modeling, and performance simulation to recommend optimized VM instance types, and dynamically adjusts these recommendations based on changing workload characteristics, while also considering hardware differences across computing devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If service providers offer multiple VM instance types optimized for different use cases, then workload matching flexibility is improved, but device complexity increases

Engineering Contradiction:

Improveworkload matching flexibilityVSAvoidservice provider network complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The optimization service automatically monitors workload performance metrics and autonomously determines when to recommend VM instance type changes, eliminating the need for manual user configuration and reducing the operational complexity of managing multiple VM types

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system continuously collects performance data from workloads and uses this feedback to dynamically adjust VM instance type recommendations, creating a closed-loop system that adapts to changing conditions without increasing operational complexity

Inventive Principle:

Principle #23Feedback

2Productivity

If VM instance types are allocated different amounts of computing resources, then resource utilization efficiency is improved, but difficulty of detecting and measuring optimal allocation increases

Engineering Contradiction:

Improveresource utilization efficiencyVSAvoidoptimal allocation detection difficulty

Core Design Contradiction:

ProductivityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent replaces manual resource allocation decisions with an automated optimization service that uses machine learning algorithms and performance metrics to automatically determine optimal VM instance type allocations, eliminating the need for complex manual analysis and measurement

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Ease of operation

If new users are provided with VM instance selection flexibility, then ease of operation is improved, but loss of time in selecting appropriate instances increases

Engineering Contradiction:

ImproveVM instance selection easeVSAvoidinstance selection time

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The optimization service performs preliminary analysis of workload characteristics and pre-determines suitable VM instance type recommendations before users need to make selection decisions, saving users time by providing ready-made recommendations based on automated analysis of their specific workload requirements

Inventive Principle:

Principle #10Preliminary action

4Productivity

If the optimization service continuously monitors workload performance, then productivity is improved, but use of energy increases

Engineering Contradiction:

Improveworkload optimization efficiencyVSAvoidoptimization service energy consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The optimization service implements periodic monitoring of workload performance metrics at scheduled intervals rather than continuous monitoring, reducing energy consumption while still maintaining the ability to detect significant performance changes and optimize VM instance type allocations effectively

Inventive Principle:

Principle #19Periodic action

Data Source

PatentUS12135980B2Compute platform optimization over the life of a workload in a distributed computing environment

Publication Date: 2024.11.05 AMAZON TECH INC

AI summary

Techniques for an optimization service of a service provider network to help optimize the selection, configuration, and utilization, of virtual machine (VM) instance types to support workloads on behalf of users. The optimization service may implement the techniques described herein at various stages in a life cycle of a workload to help optimize the performance of the workload, and reduce underutilization of computing resources. For example, the optimization service may perform techniques to help new users select an optimized VM instance type on which to initially launch their workload. Further, the optimization service may monitor a workload for the life of the workload, and determine new VM instance types, and/or configuration modifications, that optimize the performance of the workload. The optimization service may provide recommendations to users that help improve performance of their workloads, and that also increase the aggregate utilization of computing resources of the service provider network.