High-availability Grand Master Clock for IEEE 1588 Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The IEEE 1588 standard for precision clock synchronization in networked measurement and control systems faces issues with precision degradation and inconsistent grand master clock selection due to sequential redundancy, particularly in scenarios where the re-election protocol is slow and may lead to continuous re-election of faulty grand masters.
Innovation Solution
Implementing a high-availability grand master clock system using parallel redundancy, where multiple grand master clocks operate simultaneously, eliminating the need for re-election and ensuring continuous synchronization by configuring bi-directional communication links and using fault-tolerant clock synchronization methods like TTEthernet compression to maintain synchronized time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single grand master clock is used according to IEEE 1588 standard, then the system structure is simple and easy to manage, but the precision degrades and reliability decreases when the grand master clock fails or needs re-election
Solution Approach 1:
The grand master clock function is segmented into multiple independent instances (first grand master clock and second grand master clock) that operate in parallel. Each instance can independently provide synchronization services, and the system divides the single-point-failure risk across multiple segments, ensuring continuous operation even if one instance fails.
Solution Approach 2:
Different grand master clock instances are assigned different priority levels (first priority for the first grand master clock, second priority for the second grand master clock). This local differentiation in quality/priority allows the system to maintain a clear hierarchy while providing redundancy, with higher-priority clocks serving as primary sources and lower-priority clocks as backups.
2Reliability
If sequential redundancy with re-election protocol is implemented, then fault tolerance is provided, but the precision degrades during re-election and downtime occurs
Solution Approach 1:
The second grand master clock is pre-configured and ready to provide synchronization services before any failure occurs. Instead of waiting for a failure and then re-electing a new grand master, the system has a pre-prepared backup that can immediately take over, eliminating the re-election delay and ensuring continuous synchronization without downtime.
Solution Approach 2:
Both grand master clock instances operate simultaneously and continuously provide synchronization services to their respective slave clocks. The first grand master clock serves clocks that require highest precision, while the second grand master clock serves as an active backup and provides services to other clocks, ensuring that synchronization action continues uninterrupted even if one instance fails.
3Reliability
If multiple grand master clocks operate simultaneously with different priorities, then high availability is achieved, but the device complexity increases
Solution Approach 1:
Instead of having slave clocks actively search and select a grand master clock (which would increase complexity at the slave end), the grand master clocks themselves are configured with different priorities and actively announce their availability. The slave clocks simply receive and process the priority information from multiple grand masters, inverting the complexity burden to the more manageable grand master side where centralized control is more effective.
Data Source
AI summary
In a network based on IEEE 1588, comprising a plurality of nodes (201, 501) and a plurality of connections where each connection connects at least two nodes to allow communication between nodes including the exchange of messages according to a network protocol, the synchronization of IEEE 1588 is improved by allowing multiple grandmaster clocks (701) to operate simultaneously in the system. Thus, the re-election protocol of IEEE 1588 is made obsolete. For this, a multitude of nodes form a subsystem implementing a high-availability grand master clock (301) according to the IEEE 1588 Standard, wherein the subsystem is configured to tolerate the failure of at least one of said nodes forming said subsystem. Bi-directional communication link (401) are configured for physically connecting a IEEE 1588 Master clocks (201) and/or IEEE 1588 Slave clocks (201) to the subsystem implementing a high-availability grand master clock (301).