Containerized Workload Scheduling Through Network Affinity Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing approaches to scheduling containerized workloads in software defined data centers fail to adequately address dynamic application performance requirements and communication patterns, leading to suboptimal resource utilization and increased latencies.

Innovation Solution

Implement a containerized workload analyzer agent that monitors and analyzes communication patterns between containerized workloads to determine affinities and schedules them for co-location on the same node, using learned patterns and weights to optimize resource allocation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If containerized workloads are scheduled using existing approaches, then resource allocation is performed, but application latencies increase and resource utilization becomes suboptimal

Engineering Contradiction:

Improveresource utilizationVSAvoidapplication latencies

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary analysis of communication patterns between workloads before scheduling occurs. By monitoring and analyzing traffic characteristics in advance, the system pre-determines affinities and prepares scheduling decisions that optimize both resource utilization and latency performance before workloads are actually deployed.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors network traffic between containerized workloads and uses this feedback to dynamically adjust scheduling decisions. By analyzing actual communication patterns and traffic characteristics, the system refines its affinity-based scheduling to achieve optimal resource allocation while minimizing latencies.

Inventive Principle:

Principle #23Feedback

2Reliability

If containerized workloads are scheduled without analyzing communication patterns, then scheduling is simpler, but performance requirements are not met

Engineering Contradiction:

Improveperformance requirementsVSAvoidscheduling complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system enables workloads to self-determine their scheduling requirements by automatically generating and publishing affinity information based on their communication patterns. Workloads themselves provide the scheduling hints through their observed traffic characteristics, reducing the need for external complex analysis while ensuring performance requirements are met.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system transforms scheduling decisions from static resource allocation to dynamic affinity-based placement. By changing the scheduling parameter from simple resource availability to communication pattern analysis, the system achieves better performance while managing complexity through automated parameter transformation.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20250208889A1Containerized workload scheduling

Publication Date: 2025.06.26 VMWARE INC

AI summary

A method for containerized workload scheduling can include monitoring network traffic between a first containerized workload deployed on a node in a virtual computing environment to determine affinities between the first containerized workload and other containerized workloads in the virtual computing environment. The method can further include scheduling, based, at least in part, on the determined affinities between the first containerized workload and the other containerized workloads, execution of a second containerized workload on the node on which the first containerized workload is deployed.