Dynamic Resource Distribution for Software Programmable Network QoS
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Software programmable networks face inflexibility in configuring resources between guaranteed Quality of Service (QoS) and best-effort (BE) traffic, leading to issues with jitter and packet loss.
Innovation Solution
A control device dynamically reconfigures resource distribution between QoS and BE traffic by adjusting the number of queues and bandwidth allocation based on current and predicted demand, using a label distribution sequence to minimize jitter and packet loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a centralized control entity is used to configure network resources between QoS and BE service classes, then network resource distribution is controlled, but the configuration becomes inflexible and cannot adapt to changing traffic demands
Solution Approach 1:
The patent implements dynamic resource reconfiguration by enabling the network to automatically adjust the distribution of resources between QoS and BE service classes based on real-time traffic conditions. The system transitions from static centralized configuration to dynamic adaptive configuration, where resource allocation changes continuously to match demand patterns without manual intervention.
Solution Approach 2:
The network system performs self-configuration by automatically detecting traffic conditions and adjusting resource distribution between service classes. The control mechanism uses feedback from network state monitoring to autonomously reconfigure queues and bandwidth allocation, eliminating the need for external centralized control for each configuration change.
2Reliability
If resource distribution is statically configured between QoS and BE service classes, then network stability is maintained, but jitter and packet loss increase when traffic demand changes
Solution Approach 1:
The system implements feedback mechanisms that continuously monitor network traffic conditions and resource utilization. Based on this feedback, the control entity dynamically adjusts the distribution of queues and bandwidth between QoS and BE service classes, ensuring QoS requirements are met while adapting to changing traffic patterns to minimize jitter and packet loss.
Solution Approach 2:
The patent changes key network parameters dynamically, specifically the number of queues allocated to QoS versus BE traffic and the bandwidth distribution between service classes. These parameter adjustments are made in response to traffic conditions, allowing the system to maintain QoS reliability while reducing harmful effects like jitter and packet loss during demand fluctuations.
3Productivity
If the number of queues for QoS traffic is increased to improve QoS acceptance, then more QoS traffic can be handled, but network complexity and resource overhead increase
Solution Approach 1:
Instead of statically increasing the number of queues for QoS traffic, the system dynamically adjusts queue allocation based on actual QoS traffic demand. The number of queues is increased only when needed and reduced when demand decreases, maintaining high QoS acceptance capability while avoiding unnecessary complexity and resource overhead during periods of lower demand.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
A control device for a software programmable network including one or more nodes and supporting traffic of guaranteed quality of service, QoS, and best-effort, BE, traffic with a predetermined resource distribution between guaranteed QoS and BE service classes, comprises: a processing circuitry configured to determine a new resource distribution between guaranteed QoS and BE service classes different from the predetermined resource distribution according to a demand for QoS service, to calculate, for each node, a new configuration indicating the new resource distribution, and to calculate, for each entry node to the network among the one or more nodes, a label distribution sequence indicating one or more label distributions with associated activation timing information so as to reduce jitter or packet loss; and a transceiver configured to transmit the new configurations to respective nodes, and to transmit the label distribution sequences to respective entry nodes.