Fabric Multipathing via Dynamic Latency Calculations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data center and telecom networks face challenges in providing highly available and low-latency paths due to varying latency across different Ethernet ports and paths, despite configured path cost factors, which can lead to suboptimal data packet routing.
Innovation Solution
A system that synchronizes clocks across devices via link aggregation (LAG) ports and equal cost multipathing (ECMP) paths using Precision Time Protocol (PTP), determines transit delays for each port and path, sorts them by latency, and marks the lowest latency paths for preferential use by traffic demanding low latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If path cost factors are configured based on highest speed or available bandwidth, then network bandwidth utilization is improved, but actual latency varies dramatically across paths with the same path cost factor
Solution Approach 1:
The patent changes the routing parameter from static path cost factors to dynamic latency measurements. By continuously measuring actual latency using PTP and adjusting path selection based on real-time latency data, the system resolves the contradiction between bandwidth-based path cost configuration and actual latency variation, ensuring low-latency paths are selected even when bandwidth is not the differentiating factor.
Solution Approach 2:
The patent implements a feedback mechanism where latency measurements are continuously obtained through PTP clock synchronization and used to dynamically adjust path selection. This closed-loop feedback system monitors actual latency performance and redirects traffic away from high-latency paths, resolving the contradiction by making path selection responsive to actual latency conditions rather than relying on static bandwidth-based cost factors.
2Ease of operation
If static path cost factors are used for routing decisions, then routing simplicity is maintained, but network performance deteriorates due to inability to adapt to dynamic latency conditions
Solution Approach 1:
The patent performs preliminary latency measurements and path evaluations before making routing decisions. By pre-calculating and storing latency-based path rankings using PTP measurements, the system maintains routing simplicity while incorporating dynamic performance data, thus resolving the contradiction between static configuration simplicity and adaptive performance requirements.
Solution Approach 2:
The patent enables the network to self-adjust routing paths based on automatically collected latency measurements. The system autonomously monitors latency, updates path preferences, and redirects traffic without manual intervention, maintaining operational simplicity while improving performance reliability through adaptive behavior.
3Measurement precision
If dynamic latency measurements are implemented across all paths, then path selection accuracy is improved, but system complexity increases due to additional synchronization and measurement infrastructure
Solution Approach 1:
The patent leverages the universal applicability of PTP clock synchronization, which is already widely deployed in modern networks for time-sensitive applications. By using this existing infrastructure for dual purposes (time synchronization and latency measurement), the patent achieves precise latency measurements without adding significant complexity, as the same PTP mechanism serves both functions.
Solution Approach 2:
The patent combines latency measurement functionality with the existing PTP clock synchronization infrastructure. Instead of implementing a separate measurement system, the patent merges latency detection into the PTP message exchange process, where latency is derived from the time stamps already exchanged for clock synchronization, thereby achieving precise measurements without proportional increases in system complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach ensures that data packets are routed through the lowest latency paths, enhancing network performance and reliability by dynamically selecting paths based on actual latency rather than just path cost factors, thereby meeting modern network service level requirements.
Implementation Method 1
transit delay is precisely calculated using a PTP delay_request packet which is sent and received by a device. This transit delay is precisely equal to the latency of the path traversed by the PTP delay_request packet.
Data Source
AI summary
In one embodiment, a system for determining latency in paths includes logic integrated with and/or executable by a processor, the logic being adapted to synchronize clocks of two devices connected via two or more link aggregation (LAG) ports and/or multiple devices within paths through a network fabric, determine a transit delay for each LAG port and/or path, store the transit delay for each LAG port to a LAG structure along with an identifier for the LAG port and/or for each path to an equal cost multi-path (ECMP) structure along with an identifier of the path, sort the LAG ports according to each LAG port's transit delay and mark a LAG port having the lowest latency, and sort the paths according to each path's transit delay and mark a path having the lowest latency, wherein each path has an equal path cost factor.