Page 35

Industrial Ethernet Book 104

Technology 35 faulty. In these cases, system elements such as end nodes and bridges are still able to remain synchronized by taking the time from the redundant grandmasters. Having redundant connectivity from the end nodes and bridges to the redundant grandmaster clocks also allows the network to tolerate the loss of network links or even bridges while still maintaining a synchronized timebase. Multiple Synchronized Times There are many use cases that show the benefit of having the network concurrently support a “working clock”, used to trigger time-critical events, and a “universal clock” or “wall clock”, typically used to timestamp events. For these cases, IEEE 802.1ASrev will support multiple synchronized clocks. This enables the timestamping of events such as production data or measurements, and the synchronization of applications such as sensors, actuators and control units. 802.1CB The IEEE 802.1CB standard implements a redundancy management mechanism similar to the approaches known from HSR (High-availability Seamless Redundancy – IEC 62439-3 Clause 5) and PRP (Parallel Redundancy Protocol – IEC 62439-3 Clause 4). In order to increase availability, redundant copies of the same messages are communicated in parallel over disjoint paths through the network. An existing standard, Path Control and Reservation – IEEE 802.1Qca, defines how such paths can be set up. The redundancy management mechanism then combines these redundant messages to generate a single stream of information to the receiver(s). While the TSN working group has not yet finalized the standardization of a particular redundancy management mechanism, it is likely that it will be based on sequence numbers. Sequence numbers (and potential additional meta information) will be transported in a dedicated redundancy tag within the Ethernet frame similar to the VLAN tag. This redundancy mechanism will eliminate duplicates (i.e. redundant copies of the same message), but will likely not guarantee in-order delivery of the messages. 802.1Qcc TSN also provides mechanisms to improve existing reservation protocols such as SRP (Stream Reservation Protocol – IEEE 802.1Qat) in order to meet the requirements of industrial and automotive systems. These include support for more streams, configurable SR (Stream Reservation) classes and streams, better description of stream characteristics, support for Layer 3 streaming, deterministic stream reservation convergence, and UNI (User Network Interface) for routing and reservations. TSN configuration can be achieved statically by a network designer, or dynamically by a network service. For example, the Path Computation Element (PCE) as developed by the IETF (RFC 4655) and the corresponding Path Computation Communication Protocol (RFC 5440) could be extended not only to find routes through a network, but also to configure the communication schedules for time-aware shaping. Scheduling The configuration of large and complex networks is no simple task, but it is easily achieved using the right tools. For unsynchronized communication networks, configuration tools need to perform complex calculations in order to provide the worst-case latencies, communication jitter, and buffer requirements in the switches. Synchronized communication as defined in TSN requires scheduling, but does not suffer the same burden of network calculus. Scheduling takes into account the necessary latencies, jitter and buffer requirements and delivers provable determinism. For example, in IEEE 802.1Qbv the times of message transmissions in the end nodes and the forwarding times in the switches can be aligned to each other with respect to a network-wide synchronized time (as for example established by IEEE 802.1AS). The transmission and forwarding times of different messages need to be sufficiently offset from each other, such that queueing delays can be minimized. The definition of these various points in time for all synchronized traffic is called the communication schedule. Tools that produce such a schedule have to solve a complex search problem but remove the burden of network calculus and testing from network engineers. Those companies with experience developing such scheduling tools typically embed them in customer tool chains or offer them as specialized products. Use Cases There are a wide range of application areas requiring real-time communication where TSN can enable greater flexibility and ease of use without sacrificing deterministic performance. For example in wind turbines, deploying critical control over Ethernet helps to cut downtime and increase production efficiency. In railway applications, convergence of critical train control networks over Ethernet saves space, weight and power. Similarly in-car controls, communication can be converged over a TSN backbone network to offer a safe, yet low-cost solution for applications like autonomous driving. Let’s take a look at two use cases in a little more detail. Factory Automation In a discrete automation plant with multiple robots working on production lines, TSN will 2.2018 industrial ethernet book enable far greater operational flexibility. Today these robots are controlled locally, with limited synchronization between them, and bottlenecks for data access from beyond the factory floor. Where there is connectivity, it is either done over proprietary networks or via gateways. By removing local control functions or converging non-critical traffic in the same network, one could jeopardize the guarantees for communication of critical messages. By utilizing a TSN connection between these robots, the controls communication is guaranteed across the network even when converged with non-critical traffic, and all robots are synchronized to the same global time. This means that controls networks can be integrated with data networks, and many control functions can be centralized away from the robot cell, into a controls cloud where greater computing power can be utilized. Importantly, huge amounts of data from the robots are now also visible to higher layer networks without the need for gateways, enabling Machine as a Service (MaaS) type business models – simultaneously improving service and maintenance from machine builders and lowering capital expenditure for end users. Power Generation TSN will enable more precise optimization of plants, cheaper energy production and reduced maintenance costs. Power plants are large and complex, with new applications being added to the plant infrastructure over time. These multiple different systems must operate together to produce energy efficiently. However, there is often a wide variety of communications platforms, which only serves to further increase the system complexity. Where there are real-time communications requirements, these systems can be especially isolated from higher level networks. By using TSN Ethernet as a standard communications platform for all of the various systems, plant infrastructure can be greatly simplified. With TSN, systems requiring deterministic and reliable communication are integrated by scheduling time-critical traffic, while the higher layer protocols needed for SCADA management and security are able to run on the same network. The ease of maintaining only one network is clear, but TSN also has the advantage that network faults can be more quickly identified and fixed at the component level. Closer integration of power plant systems allows for further efficiency improvements. TSN opens up the possibility of giving cloud-based services access to real-time plant and turbine data to optimize performance and reduce the cost of production. Technology report by TTTech.


Industrial Ethernet Book 104
To see the actual publication please follow the link above