Microservice-Aware Packet Marking for QoS-Driven Network Allocation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing network service level agreements (SLAs) fail to account for the diverse quality of service (QoS) requirements and importance levels of multiple microservices within an application, leading to over-provisioning and inefficient resource use, particularly in multi-user scenarios like cloud gaming and AR/VR, due to limited interactions between customers and network providers.

Innovation Solution

A microservice-aware framework that assigns importance levels to data packets based on QoS requirements and user experience, using an Importance Marker Optimizer (IMO) to adapt network policies and optimize resource allocation, while preserving privacy and maintaining lightweight information exchange.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional SLA-based network service provision is used, then network services are provided according to defined service requirements, but diverse QoS requirements of multiple microservices cannot be differentiated leading to over-provisioning and inefficient resource use

Engineering Contradiction:
Improveresource allocation efficiencyVSAvoidnetwork resources
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent applies local quality by differentiating network resource allocation based on the specific QoS requirements of individual microservices. Each microservice is assigned an importance level (high, medium, low) and receives customized network resources accordingly, rather than uniform treatment. This enables critical microservices to receive prioritized resources while non-critical ones receive reduced resources, eliminating over-provisioning.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent segments the application into multiple microservices with distinct QoS requirements and treats them separately in the network provisioning process. By assigning individual importance levels to each microservice and managing their resource allocation independently, the system achieves fine-grained control over network resources, preventing both over-provisioning and under-provisioning.

Inventive Principle:
Principle #1Segmentation

2Productivity

If microservice-specific QoS requirements are considered, then resource allocation can be optimized, but network-application interaction complexity increases

Engineering Contradiction:
Improveresource allocation efficiencyVSAvoidnetwork-application interaction
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent extracts the QoS requirement differentiation logic from the network provisioning process and implements it through the importance level assignment mechanism. By separating the microservice identification and QoS analysis functions from the actual resource allocation, the system reduces the complexity burden on the network while still achieving optimized resource distribution.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces an intermediary mechanism (the importance level assignment system) that mediates between microservice QoS requirements and network resource allocation. This intermediary layer translates diverse QoS requirements into simplified importance levels (high, medium, low), which then guide network resource distribution, reducing direct interaction complexity.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If importance levels are assigned to all data packets, then QoE optimization is achieved, but information exchange between application and network increases

Engineering Contradiction:
Improvequality of experienceVSAvoidinformation exchange overhead
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The patent applies partial action by assigning importance levels only to data packets from microservices that have been identified and classified, rather than treating all packets uniformly. This selective approach optimizes QoE for critical microservices while avoiding unnecessary information processing for non-critical traffic, reducing overall information exchange overhead.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS20250279966A1Microservice-aware framework for network-application collaboration
Publication Date: 2025.09.04 NOKIA SOLUTIONS & NETWORKS OY
  • US20250279966A1 patent drawing
  • US20250279966A1 patent drawing
  • US20250279966A1 patent drawing

AI summary

Method comprisingreceiving, for each of plural microservices, a respective requirement of a quality of service, wherein the plural microservices jointly build an application;receiving a first label, wherein the first label indicates a policy status of a communication system conveying data packets from the microservices for the application;receiving an indication of a first quality of experience of the application;determining, for a first one of the microservices, an importance level based on the requirement of the quality of service of the first one of the microservices, the first label, and the first quality of experience,marking a data packet originated from or designated to the first one of the microservices and to be conveyed by the communication system by the importance level determined for the first one of the microservices;providing the data packet marked by the importance level to the communication system for conveying the data packet.