Microservice Workload Reassignment Using Two-Dimensional Runtime Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing workload orchestration systems in microservices environments rely on performance data from non-production environments, leading to inefficiencies and increased operational costs due to the inability to accurately predict workload runtime performance and resource needs in production environments.

Innovation Solution

Collect runtime performance data directly from production environments to generate two-dimensional predictions of workload type and call volume, enabling preemptive re-assignment of workloads to optimized node groupings based on actual performance characteristics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If workload orchestration relies on performance data from non-production environments, then deployment process is simplified, but workload runtime performance prediction accuracy deteriorates

Engineering Contradiction:

Improvedeployment process simplicityVSAvoidworkload runtime performance prediction accuracy

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The system performs preliminary actions by collecting performance data from production environments before making workload orchestration decisions. Runtime data collection agents continuously gather actual performance metrics from production workloads, and this data is stored in databases for later analysis. By preparing the performance data in advance through continuous collection and storage, the system can make accurate orchestration decisions without complicating the deployment process.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system introduces intermediary components including runtime data collection agents that bridge between production workloads and the orchestration system, databases that mediate data storage and retrieval, and machine learning models that act as intermediaries between raw performance data and orchestration decisions. These intermediaries enable accurate performance prediction while keeping the overall system architecture manageable.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Device complexity

If traditional workload orchestration is used without predictive analytics, then system complexity is reduced, but resource allocation efficiency deteriorates

Engineering Contradiction:

Improveorchestration system complexityVSAvoidresource allocation efficiency

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The orchestration system is segmented into distinct functional modules: runtime data collection agents that gather performance metrics, databases that store collected data, machine learning models that analyze patterns and predict future performance, and workload placement decisions that execute orchestration actions. This segmentation allows each component to perform its specific function independently, managing complexity while enabling sophisticated predictive analytics for improved resource allocation efficiency.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system implements dynamic workload orchestration by continuously collecting runtime performance data and using machine learning models to adapt to changing workload patterns. The orchestration decisions are not static but dynamically adjusted based on real-time performance predictions, allowing the system to optimize resource allocation efficiency while maintaining manageable complexity through automated adaptive behavior.

Inventive Principle:

Principle #15Dynamics

3Loss of time

If workload re-assignment is performed without accurate runtime predictions, then response time is reduced, but resource utilization optimization deteriorates

Engineering Contradiction:

Improveworkload re-assignment response timeVSAvoidresource utilization optimization

Core Design Contradiction:

Loss of timeVSProductivity

Solution Approach 1:

The system performs preliminary analysis by continuously collecting and storing runtime performance data, and by pre-training machine learning models to predict workload behavior patterns. This preliminary preparation enables the system to make rapid workload re-assignment decisions based on pre-computed performance predictions, achieving fast response times while maintaining optimal resource utilization through data-driven insights.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20260037292A1Preemptive workload re-assignment within a microservices environment in a development pipeline via two-dimensional runtime predictions

Publication Date: 2026.02.05 DISH NETWORK TECHNOLOGIES INDIA PTE LTD

AI summary

A microservices cloud environment features improved workload orchestration based on enhanced runtime performance predictions. The present disclosure provides solutions that can be implemented in a microservices development pipeline (e.g., a continuous integration and continuous delivery (CI/CD) pipeline), with the solutions enabling improved characterization of a true runtime performance of a microservice when deployed in a production, or user-facing, environment. Deployment of a microservice within the production environment is performed more optimally by characterizing the microservice via performance data collected specific to the production environment, and not to any other environment in the CI/CD pipeline. The environment-specific characterization of a microservice is based on predictions that link both workload type (e.g., compute-intensive, memory-intensive, network-intensive) and call volume. According to its characterization, a microservice can be re-assigned, after its initial deployment in a non-optimized node group, to an optimized node group in anticipation of the microservice's predicted peak call time.