Software-Defined Latency Networking via Flow Bundling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current communication systems face challenges in enforcing tight end-to-end latency guarantees across large-scale networks due to complexity, poor scalability, and underwhelming performance of existing networking tools, particularly in supporting low-latency services required by new interactive applications and 5G networks.
Innovation Solution
The implementation of a software-defined guaranteed-latency networking (SD-GLN) framework that leverages programmability at network ingress nodes to enforce tight latency bounds without requiring new hardware or signaling protocols, using techniques such as flow bundling, periodic service sequences, and credit counters to manage transmission timeslots, ensuring efficient resource utilization and latency compliance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing networking tools (IEEE TSN, IETF DetNet) are used to enforce latency guarantees, then latency bounds can be maintained, but device complexity and scalability deteriorate
Solution Approach 1:
The patent replaces complex hardware-based timing mechanisms with software-based flow bundling and timeslot allocation. By aggregating multiple flows into bundles and assigning them to timeslots in software at ingress nodes, the system achieves latency guarantees without requiring synchronized hardware timers or complex networking hardware modifications throughout the entire network path.
Solution Approach 2:
The patent introduces flow bundles as an intermediary structure between individual flows and network resources. Instead of managing each flow independently through complex hardware mechanisms, flows are grouped into bundles that share common timeslots, simplifying the intermediary layer between traffic classification and actual transmission.
2Reliability
If existing networking tools are used to enforce latency guarantees, then latency bounds can be maintained, but scalability deteriorates
Solution Approach 1:
The patent merges multiple individual flows into flow bundles that share common timeslots. This consolidation reduces the number of separate latency enforcement mechanisms needed across the network, allowing the system to scale to more flows and nodes without proportionally increasing complexity. Multiple flows are combined into bundles based on their latency requirements and path characteristics.
Solution Approach 2:
The patent creates a universal flow bundle structure that can accommodate multiple flows with different characteristics as long as they share common path segments and latency requirements. This multi-functional bundle approach allows a single timeslot allocation mechanism to serve multiple purposes, enhancing scalability across diverse network topologies and traffic patterns.
3Productivity
If flow bundling is implemented to improve scalability, then network utilization improves, but measurement precision requirements increase
Solution Approach 1:
The patent implements feedback mechanisms where ingress nodes monitor actual flow characteristics and adjust bundle formation and timeslot allocation accordingly. By continuously measuring packet arrival times, durations, and network conditions, the system refines its bundle assignments to optimize utilization while maintaining latency guarantees, balancing productivity gains with measurement accuracy requirements.
Data Source
AI summary
Various example embodiments for supporting guaranteed-latency networking are presented herein. Various example embodiments for supporting guaranteed-latency networking may be configured to support guaranteed-latency networking based on use of a time division multiplexing (TDM) frame, such as a periodic service sequence (PSS), to support transmission of packets of a set of flows. Various example embodiments for supporting guaranteed-latency networking based on use of a PSS may be configured to support guaranteed-latency networking based on use of various PSS computation enhancements. Various example embodiments for supporting guaranteed-latency networking based on use of a PSS may be configured to support guaranteed-latency networking based on use of flow bundling.