Queue Control Packet Mitigates Container Deletion Packet Loss

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In elastic container-based networks, packet loss occurs when containers are deleted during the process of managing and replicating network packets, as in-flight packets may be lost due to the destruction of containers without proper handling.

Innovation Solution

The introduction of a Queue Control Packet (QCP) is used to manage and redirect packet flows, ensuring that packets are processed and transferred to secondary containers before the primary container is deleted, thereby preventing packet loss. The QCP is injected into the data path, allowing for the safe deletion of containers without losing any packets in transit.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If containers are deleted during network packet management, then network resource utilization is improved, but packet loss occurs

Engineering Contradiction:

Improvenetwork resource utilizationVSAvoidpacket delivery reliability

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system performs preliminary actions by sending a queue control packet to the container before deletion to drain the packet queue. This ensures that all in-flight packets are processed or forwarded before the container is removed, preventing packet loss while enabling container deletion for resource optimization

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

A queue control packet is introduced as an intermediary mechanism between the container deletion process and the packet queue. This control packet mediates the transition by signaling the container to empty its queue, thus ensuring packets are handled properly during the container lifecycle management

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If packets are drained from container queue before deletion, then packet loss is prevented, but network throughput is reduced

Engineering Contradiction:

Improvepacket delivery reliabilityVSAvoidnetwork throughput

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The draining process is implemented as a periodic or staged action rather than a continuous block. The queue control packet initiates a controlled draining process that allows packets to be processed at a manageable rate, balancing reliability requirements with throughput maintenance

Inventive Principle:

Principle #19Periodic action

3Reliability

If route is stalled after sending queue control data, then packet loss is prevented, but network latency increases

Engineering Contradiction:

Improvepacket delivery reliabilityVSAvoidnetwork latency

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The route stalling is a preliminary anti-action taken to prevent the harmful effect of packet loss. By temporarily halting the route after sending the queue control packet, the system prevents new packets from being sent to a container that is in the process of being deleted, thus avoiding packet loss while minimizing latency through controlled timing

Inventive Principle:

Principle #9Preliminary anti-action

Data Source

PatentUS10476783B2Packet loss mitigation in an elastic container-based network

Publication Date: 2019.11.12 CISCO TECHNOLOGY INC

AI summary

Packet loss mitigation may be provided. First, queue control data may be sent to a first container and then a route may be stalled after sending the queue control data. The route may correspond to a data path that leads to the first container. Next, modified queue control data may be received from the first container and the first container may be deleted safely with empty queues, preventing packet loss in response to receiving the modified queue control data.