Dynamic Shared Buffer Allocation for Packet Latency Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication networks face challenges in accurately dimensioning buffer capacity in transport nodes, particularly near access points where traffic variability is high due to bursty traffic patterns, leading to inefficiencies in packet handling and increased costs from limited physical memory resources.
Innovation Solution
A method for dynamically allocating shared buffer capacity in transport nodes based on network-wide traffic flow configurations, using a centralized management entity to determine optimal allocations between interfaces, considering factors like burstiness, flow identity, and network resilience to minimize packet latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If buffer capacity is increased to handle bursty traffic, then packet loss is reduced, but hardware cost increases
Solution Approach 1:
The patent implements dynamic buffer allocation where the buffer capacity assigned to each interface is not fixed but adjusts over time based on measured traffic characteristics. The system continuously monitors traffic flow patterns and modifies buffer allocations to match actual network conditions, allowing the network to handle bursty traffic effectively without permanently over-provisioning buffer capacity.
Solution Approach 2:
The system changes the parameter of buffer capacity allocation from a static configuration to a dynamic one that varies based on traffic characteristics. By measuring traffic flow parameters such as burstiness and adjusting buffer allocation accordingly, the system optimizes the balance between handling packet loss and minimizing hardware costs.
2Device complexity
If static buffer allocation is used, then device complexity is reduced, but adaptability to traffic variability worsens
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously measures traffic characteristics at each interface and uses this information to adjust buffer allocation. The measured traffic parameters feed back into the allocation algorithm, creating a closed-loop control system that adapts to changing network conditions while maintaining manageable complexity through automated decision-making.
Solution Approach 2:
The buffer allocation system serves itself by automatically measuring traffic characteristics and adjusting its own configuration without external intervention. Each transport node autonomously monitors its interfaces and reallocates buffer capacity based on observed traffic patterns, eliminating the need for complex external management while improving adaptability.
3Quantity of substance
If shared buffer capacity is used across multiple interfaces, then hardware cost is reduced, but measurement precision for individual interfaces worsens
Solution Approach 1:
The patent applies local quality by measuring and analyzing traffic characteristics specifically at each individual interface rather than using a single global measurement. Each interface's buffer requirements are determined based on its own traffic patterns, allowing the system to allocate shared buffer capacity with precision tailored to local conditions at each interface.
Solution Approach 2:
The system segments the measurement process by interface, evaluating traffic characteristics independently for each interface before aggregating the results for shared buffer allocation. This segmentation allows the system to maintain total buffer capacity efficiency while achieving interface-specific measurement precision through separate evaluation of each interface's needs.
Data Source
AI summary
A method (200) for controlling shared buffer capacity in a packet switched network is disclosed, the packet switched network comprising a plurality of transport nodes, each node comprising a shared buffer capacity which may be allocated to interfaces of the transport node. The method comprises obtaining a configuration of traffic flows within the network (210). The method further comprises, for a transport node, determining, on the basis of traffic flows within the network, an allocation of the transport node's shared buffer capacity between interfaces of the transport node to limit overall packet latency for the transport node (220), and causing the determined allocation of shared buffer capacity to be applied in the transport node (230). Also disclosed are an apparatus (600, 800) and a transport node (700, 900) in a packet switched network and a computer program product configured to carry out methods in a packet switched network and a transport node.