Bandwidth Scheduling for Data Center Traffic Peaks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In public cloud data centers, physical lines in the data center interconnection network experience low utilization due to significant differences between traffic peaks and valleys, leading to wasted resources and potential packet loss during peak times.

Innovation Solution

A bandwidth scheduling method that allocates bandwidth based on historical data to efficiently manage non-real-time traffic, balancing resource usage by predicting and scheduling bandwidth requirements across different time periods, ensuring stable occupation of total bandwidth and reducing the difference between peak and valley traffic.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If physical line bandwidth is expanded based on peak traffic, then packet loss during transmission is avoided, but physical line resources are wasted due to low utilization during traffic valleys

Engineering Contradiction:
Improvepacket loss preventionVSAvoidphysical line resource waste
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The system performs preliminary bandwidth allocation by predicting future bandwidth requirements based on historical data and traffic patterns. The bandwidth scheduling module allocates bandwidth in advance for non-real-time traffic during traffic valleys before peak periods occur, ensuring that physical lines are fully utilized without requiring expansion for peak traffic alone.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts bandwidth allocation based on real-time traffic conditions and predicted future needs. The bandwidth scheduling module continuously monitors traffic patterns and modifies allocation strategies, transitioning from static peak-based expansion to dynamic allocation that adapts to varying traffic demands throughout different time periods.

Inventive Principle:
Principle #15Dynamics

2Reliability

If physical line bandwidth is expanded to handle peak traffic, then transmission reliability is improved, but device complexity increases due to managing expanded capacity

Engineering Contradiction:
Improvetransmission reliabilityVSAvoidphysical line management complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system implements feedback mechanisms where the bandwidth scheduling module continuously receives traffic status information, analyzes patterns, and adjusts bandwidth allocation accordingly. This closed-loop control enables the system to maintain transmission reliability through intelligent scheduling rather than physical expansion, reducing the complexity of managing expanded infrastructure.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system changes operational parameters by shifting from fixed bandwidth allocation based on peak traffic to variable allocation based on predicted traffic patterns. The bandwidth scheduling module adjusts allocation parameters dynamically, allowing the same physical infrastructure to handle varying loads efficiently without requiring additional physical lines or complex management of expanded capacity.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If bandwidth is allocated based on historical information for non-real-time traffic, then physical line utilization is improved, but bandwidth allocation complexity increases

Engineering Contradiction:
Improvephysical line utilizationVSAvoidbandwidth allocation complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The bandwidth scheduling module operates autonomously by automatically analyzing historical bandwidth information, predicting future requirements, and allocating bandwidth without manual intervention. The system serves itself by continuously optimizing allocation based on learned patterns, improving physical line utilization while managing allocation complexity through automation rather than manual processes.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system replaces manual or mechanical bandwidth allocation methods with automated computational analysis. The bandwidth scheduling module uses algorithms to process historical data and generate allocation decisions, substituting complex manual planning with automated systems that handle the complexity internally while presenting simple interfaces for resource management.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4072086B1Bandwidth scheduling method, traffic transmission method, and related product
Publication Date: 2024.08.07 HUAWEI CLOUD COMPUTING TECHNOLOGIES CO LTD
  • EP4072086B1 patent drawingFigure 1
  • EP4072086B1 patent drawingFigure 2~3
  • EP4072086B1 patent drawingFigure 4~5

AI summary

Embodiments of this application disclose a bandwidth scheduling method, a traffic transmission method, and a related product. The bandwidth scheduling method includes: receiving a bandwidth request sent by a data center, where the bandwidth request includes bandwidth required to transmit non-real-time traffic; allocating, based on historical bandwidth information, the bandwidth required by the data center to transmit the non-real-time traffic in a future time period, where the historical bandwidth information is used to predict occupation of total bandwidth by the data center in a region in which the data center is located at each moment in the future time period; and sending a bandwidth response to the data center, where the bandwidth response includes an allocation result. Embodiments of this application help improve utilization of physical line resources in each region.