A Deep Dive into Time-Sensitive Networks
Current industry trends such as Industry 4.0 and the Industrial Internet of Things (IIoT) are leading to more communication in ever-growing converged networks. Such networks require flexibility and scalability to support everything from small devices to machinery and production line control devices, as well as large data server systems. They must also guarantee limited latency for critical real-time communications. Time-Sensitive Networking (TSN) is defined by the IEEE, the Ethernet Standards Organization. It is intended to cover all these requirements to allow the simultaneous use of deterministic and non-deterministic communications in converged networks.
Before the features of TSN can be detailed, it should first be clarified that TSN is not a protocol. Instead, TSN is an umbrella term for a set of features that collectively enable standardized deterministic Ethernet. Without these features, standard Ethernet is not able to operate in real time. Note: The TSN features described below were chosen based on the IEC/IEEE 60802 Common Profile for Industrial Automation.
Time synchronization and prioritization
Two long-established principles in Ethernet underlie TSN functionality: time synchronization and prioritization. First, TSN works best if the internal clock of each transmitter and receiver is synchronized with other clocks on a network. Second, data flowing over a network from a sender to a receiver is called a flow and is assigned a priority. IEC/IEEE 60802 specifies four priority classes. For our purposes, let’s simplify them into just three classes: high, low, and best effort. If we take the vehicles on a roadway by analogy, the prioritization would be similar to the lanes reserved for HOVs on a highway.
Besides senders and receivers, the other key piece of hardware in any network is the Ethernet switch or, in TSN parlance, a bridge. The characteristics of a bridge can also be explained using the analogy of roads and vehicles. Bridges are best thought of as roundabouts. Vehicles enter the roundabout (bridge), cross the circle if necessary and exit via the appropriate carriageway. Similarly, Ethernet frames arrive at a bridge and are directed to the correct port without loss of congestion. This prevention of loss of congestion at bridges is one of the critical elements of TSN.
Time-conscious planning and shaping
Think of shippers as a factory at the end of a shift. All the workers want to leave at exactly the same time. To avoid loss of congestion (and maintain limited latency) at the bridge, senders schedule the order in which they transmit their frames based on priority.
Once at the bridge, prioritization alone is not the only method of determining which vehicles can exit at the desired point. Internally, bridges can configure similar priority frame queues. The algorithms would then determine the next frame per queue to send. This would imply the possibility that queues with lower priorities would never be sent. Instead, priority queues are replaced with a repeating cycle of time slots to ensure that frames of all priorities can be sent. This is known as a time-aware shaper (TAS) and is analogous to a city bus timetable. Each time slot begins with high priority frames.
Then come the low priority frames. Finally, the time remaining in the slot is filled with best effort frames that do not rely on time awareness. This way, all priorities are passed while maintaining determinism for high priority traffic.
At the end of each time slot, the TAS creates a short time window called the guard band. Its purpose is to ensure that the transmission of best effort frames does not overflow onto high priority frames in the next slot, and therefore does not delay them. Often, best effort frameworks are relatively large. In this case, the guard band must be as large as the transmission time for the potentially largest frame.
TAS is inefficient with its bandwidth in this way. Instead, time-sensitive bridges can calculate the transmission time for these large frames as best they can and then decide whether or not they can complete their transmission before the end of the slot. Otherwise, these frames are broken into fragments, transmitted separately, and recombined at the next bridge. This procedure is called preemption. It minimizes guard band size, eliminates latency and congestion loss, and maximizes time slot utilization (i.e., best effort traffic throughput).
Limits of RST
In converged networks, it is likely that one TSN domain will be connected to non-TSN enabled devices, or even to another TSN domain. Ports on bridges connected to devices outside the TSN domain are called boundary ports. To protect network resources inside the domain, boundary ports invoke a feature known as ingress rate limiting.
If we go back to vehicles on a highway analogy, the entrance rate limiter works like a traffic light on the freeway ramp. The traffic light regulates vehicle access to the freeway depending on the traffic density on the freeway.
However, these TSN limits are smarter than your typical traffic light. They can individually assign priorities, allowing vehicles to join HOV lanes directly. In technical terms, this is the remapping of VLAN priorities and assignments.
Network management engine
If this all sounds awfully complex, that’s because it is! Even Ethernet today, if you were intimately aware of its inner workings, is quite complex. And yet, Ethernet is incredibly easy to use. The same goes for those complex TSN features. The software that enables and manages these mechanisms, in accordance with IEC/IEEE 60802, is the TSN Domain Management Entity (TDME). Within Profinet, we call this the Network Management Engine (NME). The NME takes care of all the calculations, planning, configuration and resource allocation to make administration of a TSN network easy or even invisible to a user.
You may be wondering: why do all this? Doesn’t such functionality already exist in some industrial Ethernet protocols (eg Profinet)?
The answer is yes, but we’re not here for the features, we’re here for the benefits they provide. For example, due to IEEE standardization, a wide variety of hardware is available with off-the-shelf TSN compatible Ethernet chips. This is beneficial to automation component vendors as it reduces costs and stimulates competition among technology vendors.
For end users, the benefits are clear: TSN-based converged networks enable out-of-the-box manufacturing. New concepts of machines and networks are now made possible by this technology.