Service Mesh Latency Prediction via MUE Ratios

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In Cloud Native environments, it is challenging to increase the operational efficiency of a service mesh while reducing costs by dynamically determining the optimal allocation of resources such as CPU compute power, memory, and networking bandwidth to meet service level agreements (SLAs), as adjusting these resources may not necessarily improve operational efficiency.

Innovation Solution

The introduction of Service Mesh Utilization Efficiency (MUE) values, calculated using a combination of metrics and key performance indicators (KPIs), allows for the determination of resource utilization and optimization of latency and throughput without direct measurement, enabling the adjustment of resource allocation to achieve desired service level objectives.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If resources (CPU, memory, networking bandwidth) are dynamically allocated to service mesh, then operational efficiency should improve, but resource allocation adjustments may not necessarily improve operational efficiency

Engineering Contradiction:

Improveservice mesh operational efficiencyVSAvoidservice level agreement fulfillment

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system implements a feedback mechanism by continuously monitoring service mesh performance metrics (latency, throughput, error rates) and using this information to dynamically adjust resource allocation. The feedback loop ensures that resource adjustments are based on actual performance data, resolving the contradiction by making resource allocation responsive to real operational conditions rather than static configurations.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent applies dynamics by transitioning from static resource allocation to dynamic resource allocation that adapts to changing workload conditions. The system continuously monitors performance metrics and adjusts CPU, memory, and networking bandwidth allocation in real-time, allowing the service mesh to optimize operational efficiency while maintaining service level agreements under varying load conditions.

Inventive Principle:

Principle #15Dynamics

2Reliability

If resources are increased to meet service level agreements, then service level agreement fulfillment improves, but costs increase

Engineering Contradiction:

Improveservice level agreement fulfillmentVSAvoidresource allocation

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The system changes parameters by dynamically adjusting resource allocation levels based on monitored performance metrics and predicted future conditions. Instead of maintaining fixed high resource allocation to ensure service level agreements, the system varies CPU, memory, and networking bandwidth allocation according to actual demand and predicted workload patterns, reducing resource consumption while maintaining agreement fulfillment.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent applies preliminary action by using machine learning models to predict future workload conditions and performance metrics. This predictive capability allows the system to proactively adjust resource allocation before performance degradation occurs, ensuring service level agreements are met while avoiding unnecessary resource allocation during periods of low demand.

Inventive Principle:

Principle #10Preliminary action

3Productivity

If resource allocation is adjusted dynamically, then operational efficiency can be optimized, but system complexity increases

Engineering Contradiction:

Improveservice mesh operational efficiencyVSAvoidresource management system complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system introduces an intermediary layer consisting of machine learning models and performance monitoring components that mediate between workload conditions and resource allocation decisions. This intermediary automatically analyzes performance metrics, predicts future conditions, and determines optimal resource allocation, reducing the complexity burden on operators while enabling dynamic optimization of service mesh performance.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20230105491A1Cloud service mesh performance tuning

Publication Date: 2023.04.06 INTEL CORP

AI summary

Examples described herein relate to a system to estimate latency of operations of a process without receiving a latency value directly based on received performance values and/or estimate throughput of packets transmitted for the process without receiving a throughput value directly based on received performance values. In some examples, the system is to request to adjust resource allocation to perform the process based on the determined latency and throughput.