Network Node Resource Allocation via Queue and Flow Rate Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current cloud computing and network processing systems face challenges in dynamically managing resource allocation and deallocation across interconnected hardware infrastructure, leading to potential congestion and delays in packet flow due to inefficient queuing and resource utilization.

Innovation Solution

A physical network node is implemented to determine the number of requests and packet flow rate across serially-connected service nodes, allowing for controlled allocation or deallocation of resources at a future time based on these conditions, thereby optimizing resource usage and preventing congestion.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If resources are dynamically allocated and deallocated to service nodes based on current network conditions, then resource utilization efficiency is improved, but packet flow delays may occur due to allocation timing mismatches

Engineering Contradiction:
Improveresource utilization efficiencyVSAvoidpacket flow delays
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by determining future allocation times based on current queue lengths and packet flow rates. The network node calculates when resource allocation should occur in advance, before the actual allocation event, to prevent congestion before it happens. This proactive approach ensures resources are available when needed without causing delays.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts resource allocation timing based on real-time network conditions including queue lengths and packet flow rates. The allocation strategy adapts to changing traffic patterns, service node processing capacities, and network demand, optimizing resource utilization while preventing congestion through continuous adjustment of allocation decisions.

Inventive Principle:
Principle #15Dynamics

2Reliability

If resource allocation is based on current queue length and packet flow rate, then congestion is prevented, but system complexity increases due to multiple parameters to monitor and calculate

Engineering Contradiction:
Improvecongestion preventionVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system implements feedback mechanisms by continuously monitoring queue lengths at service nodes and packet flow rates, using this information to determine optimal resource allocation timing. The network node adjusts allocation decisions based on feedback from network conditions, ensuring congestion prevention while managing complexity through structured feedback loops.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The network node autonomously determines resource allocation timing by self-monitoring network conditions and automatically calculating future allocation times based on queue lengths and packet flow rates. This self-service approach reduces the need for external control mechanisms while maintaining reliable congestion prevention through intelligent autonomous decision-making.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3420452B1Interconnected hardware infrastructure resource control
Publication Date: 2020.11.04 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
  • EP3420452B1 patent drawingFigure 1
  • EP3420452B1 patent drawingFigure 2
  • EP3420452B1 patent drawingFigure 3A

AI summary

A physical network node controls allocation and/or deallocation of resources of an interconnected hardware infrastructure. The physical network node determines a number of requests currently queued at a first service node of a plurality of serially-connected service nodes at a current time. The plurality of serially-connected service nodes supports a packet flow using resources of the interconnected hardware infrastructure. The physical network node also determines a packet flow rate of the packet flow into the first service node. The physical network node also determines a future time to control allocation or deallocation of a resource of the interconnected hardware infrastructure to a second service node of the plurality of serially- connected service nodes based on the determined number of requests and the determined packet flow rate. The physical network node controls allocation or deallocation of the resource to the second service node at the future time.