The easier way to sync up data networks

Supporting precise synchronization is essential to digital transformation, but until recently it was a major cost and complexity challenge. Time to look at the new technologies enabling network operators across all industries to integrate sync into their existing hardware platforms.
Arthur Cole

Data networks are getting faster, volumes are increasing and the number of interdependent services is on the rise. For network operators, this means timing and synchronization are becoming just as vital to overall performance as throughput and bandwidth.

When we think of all the devices that are becoming digitized, the need for highly accurate timing becomes evident. Autonomous cars will need to communicate with traffic signals, GPS services, and each other, in order to function properly. Medical devices will be less reliable and provide only minimal services if they cannot access the right data at the right time. Even emerging services like Fintech and virtual reality must perform precise calculations based on data from external sources in order to meet customer expectations.

And Facebook can forget about the metaverse if it cannot support timing and synchronization down to the millisecond, of even less, across a wide swatch of supporting technologies.

Integrated timing

To date, however, timing has been something of a forgotten child in network circles. Yes, it is highly valuable, but most platform developers add it as an external function in key switching and routing infrastructure. This is largely due to the fact that there are few options when it comes to integrating timing and sync into hardware platforms.

The situation becomes even more difficult as highly specialized network and data center equipment is replaced by white box switches and servers using software to provide essential network functions. While most original device manufacturers (ODM) understand that precise time needs to be provided in close proximity to applications running on servers in the data center, they don’t have the necessary sync expertise.

What this offers is a chance for ODMs and synchronization experts to join forces for a win-win approach. 

A new generation of internal timing modules is emerging to allow ODMs to integrate highly precise synchronization capabilities into their product lines. Among the first to hit the market is the OSA 5400 SyncModule™, which ADVA has already integrated into its Ensemble Controller to enable comprehensive GNSS and PTP assurance.

In the past, equipment vendors had to incorporate sync and timing using chip-level components from specialty suppliers and  develop their own solutions in-house adding considerable cost and complexity to designs. Solutions like the OSA 5400 provide a shortcut around these problems, providing both satellite- and network-based timing along with higher levels of assurance and control.

As the 5G era unfolds, we can expect synchronization to become even more exact.

Open and connected

M.2-based solutions are expected to be particularly helpful to open compute platforms, which often lack the precise timing of proprietary solutions due to their tendency to compile diverse systems from multiple vendors. However, as more of the wired and wireless communications infrastructure comes to rely on these white-box platforms, tighter data coordination than is typically available in a common API-based solution will become paramount. In many cases, sub-microsecond timing across public and private networks will be necessary, particularly those that support critical workloads.

Data center and mobile networks are also converting to open platforms, and their need for accurate timing is growing as well. Since much of this infrastructure already incorporates the PCIe bus, ADVA has released the OSA 5400 TimeCard™ to embed PTP and NTP support onto any Open Compute Project server or OpenRAN device, instantly turning it into a grandmaster, boundary or slave clock. This should greatly enhance application sync by harnessing data center, access and edge network infrastructure under the same time framework.

We can also expect synchronization to become even more exact as the 5G era unfolds. Astra Baltayan, product marketing manager at network analysis firm Exfo, noted recently that numerous elements across the entire radio access network might soon require sync of better than  100 nanosecond. With advanced capabilities like dynamic spectrum sharing (DSS) and massive multiple-input multiple-output (MIMO) permeating an increasing number of cell towers and user equipment, proper handoff from one device to another will be crucial to meet the demands of an increasingly digitized economy.

The value of today’s network environment lies in its ability to connect disparate resources so they function as an integrated data system. The more data it contains, and the faster it moves, the greater the need for highly synchronized operations. By deploying sync and timing capabilities onto existing form factors, network operators can quickly integrate legacy infrastructure to support next-generation data services.

Related articles