Synchronous Ethernet Network Architecture and Configuration

Synchronous Ethernet is a technology that allows for the synchronization of Ethernet networks with a precise clock signal. This is achieved through the use of a synchronization mechanism that is integrated into the Ethernet network.

The IEEE 1588 standard is a key component of Synchronous Ethernet, providing a framework for synchronizing clocks across the network. This standard is widely adopted in the industry.

In a Synchronous Ethernet network, the clock signal is distributed throughout the network using the IEEE 1588 protocol. This allows for precise synchronization of clocks across the network, even over long distances.

Synchronous Ethernet networks can be configured using a variety of methods, including the use of Network Time Protocol (NTP) and the IEEE 1588 protocol.

The Renesas 8A34001 Synchronization Management unit (SMU) chip is used as a Clock Synthesizer/Jitter Cleaner for the GT_Ref_Clock, GT recovered clock, Timer syncer Clock, and PPS signal.

For SyncE to work, a platform with specific circuitry is required. This circuitry enables the GT recovered clock generated from GT to be used as GT Ref clock for transmitting data in the downstream direction.

The platform has two Ethernet interfaces connected to two GT channels (channel-0 and channel-1). The recovered clock from each GT channel is connected to the MUX in GT primitive OBUF_DS_GTE3/4_ADV present in gt_shared IP.

An AXI Lite interface connected to gt_shared IP will write to internal DRP register (001Fh) to switch recovered clock based on syncE clock management MUX control selection API.

The synced utility is a user space application that runs on PS, interacts with the DPLLs in 8A34001 chip through the Linux driver module package to support SyncE applications.

The synced utility selects the best clock, based on Quality Level (QL) of the clock, and propagates it to the connected nodes via SyncE Ethernet Synchronization Message Channel (ESMC).

If this caught your attention, see: IP Multimedia Subsystem

Synchronization is a crucial aspect of Synchronous Ethernet. Any network element (NE) should have at least two reference clocks to ensure reliable synchronization.

Ethernet interfaces must be able to generate their own synchronization signal in case they lose their external reference, a process known as holdover. This signal must be filtered and regenerated by a phase locked loop (PLL) at the Ethernet nodes.

There are several ways to synchronize nodes in SyncE, including external timing, line timing, through timing, and internal timing. External timing is typically achieved by obtaining a signal from a stand-alone synchronization equipment (SASE).

Here are the different synchronization models in SyncE:

Synchronization Status Messages (SSM) provide a mechanism to determine the quality level of the clock sourcing a given synchronization trail and allow a network element to select the best of multiple input synchronization trails.

A SyncE chain is a series of Ethernet equipment connected in a way that allows a reference clock to be transmitted from one device to the next.

For a SyncE chain to work, all the equipment in the chain must support SyncE, as a single piece of equipment that doesn't break the chain.

The SyncE signal degrades when passing through the network, so it's filtered and regenerated by a phase-locked loop (PLL) at each Ethernet node to maintain its integrity.

If one device in the chain loses its external reference, it can generate its own synchronization signal, a process known as holdover.

Here are the different ways a SyncE chain can be broken:

In any of these cases, the downstream devices will recognize that the signal is out of pull-in range and can't use it for reference.

Synchronization relies heavily on a reliable method of aggregating multiple links, and that's where LAG comes in.

LAG, or Link Aggregation Group, allows for the bundling of multiple Ethernet links into a single logical link.

To configure synchronous Ethernet over LAG, an aggregated-ether (AE) group shall be defined as the clock source.

Any number of interfaces from the same AE group can be configured as the clock source.

The ESMC transmit shall be configured on individual link.

Synchronous Ethernet is a physical layer technology that functions regardless of the network load and supports hop-by-hop frequency transfer. It enables the delivery of synchronization services that meet the requirements of present-day mobile networks and future LTE-based infrastructures.

The ITU recommendations for Synchronous Ethernet operation are described in three documents: G.8261, G.8262, and G.8264. G.8261 defines the architecture and wander performance of Synchronous Ethernet networks, while G.8262 specifies timing characteristics of synchronous Ethernet equipment clock (EEC). G.8264 describes the Ethernet Synchronization Message Channel (ESMC).

To configure Synchronous Ethernet, you can perform the task under the [edit chassis synchronization] hierarchy. A configuration snippet is provided below:

Synchronous Ethernet, or SyncE, is the ability to provide PHY-level frequency distribution through an Ethernet port. It's a key component of Next Generation Networks (NGNs).

SyncE is designed to integrate into any existing SONET/SDH synchronization distribution architecture, allowing for easy migration from traditional to new synchronous interfaces. This makes it an attractive option for Tier 1 carriers looking to upgrade their synchronization infrastructure.

SyncE uses the physical layer of the Ethernet to transmit frequency to remote sites, making it an ideal solution for packet networks. By leveraging the physical layer, SyncE can provide timing to multiple remote network elements (NEs) through a packet network.

For another approach, see: Packet Layer Protocol

The SyncE feature relies on the Ethernet Synchronization Message Channel (ESMC) to transmit timing information. ESMC is based on an IEEE 802.3 organization-specific slow protocol and is used to convey clock information between nodes.

ESMC carries a Quality Level (QL) identifier that identifies the timing quality of the synchronization trail. QL values in QL-TLV are the same as QL values defined for SONET and SDH SSM.

Consider reading: Timing Synchronization Function

To configure Synchronous Ethernet, you'll need to enable it under the [edit chassis synchronization] hierarchy.

SyncE configuration can be performed using a specific hierarchy, which is mentioned in Example 1.

SyncE is enabled by default in European SDH networks, as shown in Example 3.

Automatic synchronization of the network clock should be enabled for SyncE configuration to work properly.

You'll need to ensure that the network-clock-select and network-clock-participate commands do not exist in the configuration to proceed with the SyncE configuration, as stated in Example 4.

To configure SyncE using ESMC and SSM, you'll need to perform a specific task, as described in Example 2.

You can display the associated show information to verify the SyncE configuration, which is shown in Example 3.

The SyncE configuration snippet is provided under the [edit chassis synchronization] hierarchy, as mentioned in Example 1.

Platform Specific Behavior is a crucial aspect to consider when configuring your network. This is because different platforms can behave differently, even when it comes to seemingly identical features.

The MX Series platforms, for instance, have some unique characteristics that set them apart from other platforms.

To confirm platform and release support for specific features, you can use the Feature Explorer. This tool will give you a clear picture of what's supported and what's not.

The ACX Series, on the other hand, has some specific requirements for Synchronous Ethernet. This feature is supported on ACX Series routers with Gigabit Ethernet and 10-Gigabit Ethernet SFP and SFP+ transceivers.

Here's a quick rundown of the platform-specific behaviors you should be aware of:

Synchronization in networks is crucial for maintaining precise timekeeping, especially in modern networks that rely on multiple topologies. A general requirement for SyncE is that any network element should have at least two reference clocks.

The most common network topologies are tree, ring, and meshed. A tree topology relies on a master clock, but it has two weak points: it depends on only one clock, and the signals gradually degrade. Ring topologies use ring configurations to propagate the synchronization signal, but care must be taken to avoid the formation of synchronizing loops.

There are several ways to synchronize nodes in SyncE, including external timing, line timing, through timing, and internal timing. In external timing, the EEC obtains its signal from a stand-alone synchronization equipment (SASE).

Synchronization signals can be analog or digital, and can be transported at various bit rates. The most common signals are 1.544 and 2.048 MHz, as well as SyncE signals at any bit rate.

A fresh viewpoint: Common Gateway Interface

Network topologies play a crucial role in synchronization networks, and understanding the different types can help you design a reliable and efficient system.

A tree topology is a basic setup that relies on a master clock, which is distributed to the rest of the slave clocks. However, this has two major weaknesses: it depends on only one clock, and the signals can degrade over time.

Ring topologies are essentially tree topologies with a ring configuration to propagate the synchronization signal. This setup offers a way to secure the tree topology, but it requires careful planning to avoid the formation of synchronizing loops.

Meshed topologies, on the other hand, have nodes that form interconnections with each other, providing redundancy in case of failure. However, synchronization loops can easily occur in these networks and should be avoided.

Here are the common network topologies used in synchronization networks:

To minimize problems associated with signal transport and avoid depending on a single clock, modern networks often combine multiple topologies and use duplication and security features, such as multiple master clocks and synchronization management protocols.

Interconnection of nodes is a crucial aspect of synchronization in networks. There are two basic ways to distribute synchronization: intranode and internode.

Intranode synchronization is achieved through a high-quality slave clock known as a synchronization supply unit (SSU). This unit is responsible for distributing synchronization to nodes situated inside the node.

The SSU is a reliable method for synchronization within a node. It ensures that all nodes within the node are synchronized.

Internode synchronization involves sending the synchronization signal to another node through a dedicated link or PHY signal. This method is used to synchronize nodes between different nodes.

Several types of networks can be used to transport the synchronous signal, including T1/E1, SONET/SDH, and SyncE. These networks can be combined to achieve synchronization.

Legacy Ethernet is not suitable for transmitting synchronization signals. If the signal crosses a legacy Ethernet island, synchronization is lost.

Take a look at this: Coding Tree Unit

Routers play a crucial role in maintaining synchronization in networks. They can support Synchronous Ethernet, a technology that ensures stable frequency synchronization to a Primary Reference Clock (PRC).

The ITU-T G.8264 specification defines the Synchronization Status Message (SSM) protocol for Synchronous Ethernet, ensuring interoperability between Synchronous Ethernet equipment. This protocol is essential for frequency transfer in networks.

To enable Synchronous Ethernet on routers, all nodes from the PRC to the last downstream node must be Synchronous Ethernet capable. This is a requirement for the technology to work effectively.

Synchronous Ethernet is particularly useful for mobile networks, such as 2G or 3G base stations, which require frequency-only synchronization. It's a recommended technology for these types of networks.

MX Series routers can support Synchronous Ethernet on the 16-port 10-Gigabit Ethernet MPC, but only if there's another MPC with an EEC present in the chassis. If not, the system will notify the user through log messages and CLI output.

Synchronization Technology is a crucial aspect of Synchronous Ethernet. There are several ways to synchronize nodes in SyncE, including external timing where the EEC obtains its signal from a stand-alone synchronization equipment.

In SyncE, external timing is a typical way to synchronize, and the Network Equipment (NE) usually also has an extra reference signal for emergency situations. This ensures a reliable and redundant synchronization system.

One of the most common methods is through timing, where the Tx outputs of one interface are synchronized with the Rx inputs of the opposite interface. This can be achieved through various means, including line timing and internal timing.

Here are the four ways to synchronize nodes in SyncE:

Synchronization signals play a crucial role in maintaining the integrity of network timing. They are the backbone of synchronization technology, ensuring that devices stay in sync with each other.

There are many signals suitable for transporting synchronization, including analog and digital signals. Analog signals come in frequencies of 1.544 and 2.048 MHz, while digital signals operate at 1.544 and 2.048 Mbit/s.

SyncE signals can be transmitted at any bit rate, making them a versatile option for synchronization. STM-n/OC-m line codes are also used for synchronization, providing a reliable method for transporting timing information.

Here are some common synchronization signals:

This chapter was initially written for SR OS Release 8.0.R7. The CLI in the current edition is based on SR OS Release 14.0.R6.

There are no software prerequisites for this configuration.

The hardware requires the use of Synchronous Ethernet capable MDA-XPs/CMA-XPs or the IMMs.

Synchronous Ethernet is only supported on optical interfaces.

Synchronization Management is a crucial aspect of Synchronous Ethernet (SyncE) that ensures the quality and reliability of the timing signal.

To ensure the quality level of the clock sourcing a given synchronization trail, Synchronization Status Messages (SSM) can be used to determine the best clock source.

SSM allows network elements to autonomously provision and reconfigure their synchronization references, preventing timing loops.

In SyncE, SSM is provided through the Ethernet Synchronization Messaging Channel (ESMC), which uses Ethernet OAM PDU to exchange Quality Level values over the SyncE link.

To configure the network for SyncE, the network clock must be enabled for automatic synchronization, and the network-clock-select and network-clock-participate commands must not exist in the configuration.

A timing loop is a completely unstable situation that can cause an immediate collapse of part of the network within the loop.

Timing loops occur when a clock signal has closed itself, but there is no master or slave clock to generate a non-deficient clock signal. This can happen when a network element (NE) is left without a reference clock, causing it to choose an alternative synchronization signal that ends up being the same signal, just returning by another route.

To prevent timing loops, ESMC allows the best clock source to be traced and defined, helping to prevent a timing loop from occurring in the first place.

The Quality Level (QL) identifier in ESMC helps identify the timing quality of the synchronization trail, which is crucial in preventing timing loops. QL values in QL-TLV are the same as QL values defined for SONET and SDH SSM.

Here are some key characteristics of timing loops:

Synchronization management is a critical aspect of Synchronous Ethernet (SyncE) configuration. To enable automatic synchronization of the network clock, you need to first configure the network clock.

The SyncE feature helps to overcome the difficulty of providing timing to multiple remote network elements (NEs) through an external time division multiplexed (TDM) circuit. It leverages the physical layer of the Ethernet to transmit frequency to the remote sites.

To configure SyncE, you need to ensure that the network-clock-select and network-clock-participate commands do not exist in the configuration. This will allow you to continue with the SyncE configuration.

SyncE uses the Ethernet Synchronization Message Channel (ESMC) to transmit synchronization information through the network. ESMC carries a Quality Level (QL) identifier that identifies the timing quality of the synchronization trail.

The ESMC channel provides the service of synchronous timing distribution, even if the Ethernet network is not required to be synchronous on all links or in all locations. This is achieved through the use of flags and TLVs (type, length, value) structure.

On a similar theme: Synchronization Channel

To achieve network synchronization, synchronization information is transmitted through the network via synchronous network connections with performance of egress clock. This is similar to the SONET/SDH synchronization network.

Here are some key restrictions to keep in mind when configuring SyncE:

In a general requirement for SyncE, any network element (NE) should have at least two reference clocks. This ensures that the Ethernet node (EN) can generate its own synchronization signal in case it loses its external reference.

Synchronization Modes are crucial in Synchronous Ethernet to ensure that data is transmitted accurately and efficiently. There are several ways to synchronize nodes in SyncE.

External timing is a common method where the EEC obtains its signal from a stand-alone synchronization equipment (SASE). This is typically done for emergency situations or as a primary synchronization source.

In line timing, the NE obtains its clock by deriving it from one of the input signals. This is a common practice in many networks.

There are also three other methods: Through timing, Internal timing, and Through timing (where the Tx outputs of one interface are synchronized with the Rx inputs of the opposite interface).

Here's an interesting read: Interface Message Processor

Synchronization models are used to ensure that nodes in a network are properly aligned. There are several ways to synchronize nodes in SyncE.

External timing is a common method where the EEC obtains its signal from a stand-alone synchronization equipment (SASE). This is often used as a backup in case the main synchronization signal is lost.

Line timing is another method where the NE derives its clock from one of the input signals. This is a simple and straightforward approach that works well in many cases.

Through timing is used when the Tx outputs of one interface are synchronized with the Rx inputs of the opposite interface. This creates a closed loop that helps maintain synchronization.

Internal timing is used when no other clock is available, such as in a temporary holdover stage after losing the synchronization signal. In this mode, the internal clock of the EEC is used to synchronize outputs.

Curious to learn more? Check out: Is Internet Faster When You Plug in Ethernet vs Wifi

Here are the different synchronization models in SyncE, summarized:

QL Selection Mode Disabled is a common configuration in North American SONET networks, where the automatic reference selection mechanisms are not used. This mode is often required when adding SyncE to such networks.

In QL Selection Mode Disabled, SyncE is enabled on MDA 1 of card 1. The synchronous interface timing can be configured with specific parameters.

The synchronous interface timing configuration parameters for the first timing reference, ref1, are set. The source port for ref1 is 1/1/2.

The synchronous interface timing for ref1 with source port 1/1/2 is configured with detailed settings.

For your interest: Asynchronous Transfer Mode