For years, routed networks and optical transport networks have been on different trajectories with respect to disaggregation and open interoperability. But common goals, technical innovations and general convergence are beginning to align these segments.
To better understand network operator views and plans for disaggregation across the transport network, Heavy Reading launched a comprehensive survey-based research project with collaboration partners Infinera, IP Infusion, Volta and Telecom Infra Project (TIP).
Focusing on disaggregation trends in routing and in optical networks, this blog is the second of two highlighting the key findings from the Open and Disaggregated Packet and Optical Networks Market Leadership Study.
Disaggregation for 5G
AT&T, China Mobile, Deutsche Telecom, NTT DoCoMo, Rakuten Mobile, Telefónica and Vodafone are just some of big name operators championing an open RAN and working with organizations like 3GPP, O-RAN Alliance, TIP and others to accomplish this goal. In doing so, these operators expect faster innovation, lower costs and a break from vendor lock-in.
Functional decomposition of the radio unit (RU), distributed unit (DU) and central unit (CU) RAN components is a prerequisite for RAN interoperability. Such decomposition presents operators with multiple architectural options, each of which affects transport network technologies and choices. On one end is the fully distributed RAN (like traditional macro cells). On the other end is a fully centralized RAN. And, in between, there are several other split RAN variations.
In the survey, Heavy Reading asked operators to estimate the 5G RAN functional splits options they expect to deploy in their emerging 5G RANs over the next three years. The figure below shows the average functional split breakdown for the full survey group.
While there is no single standout among the options, the results are a strong endorsement for split architectures generally. The traditional distributed macro architecture will account for roughly a one-quarter share of emerging 5G RAN based on the results, but nearly two-thirds (63%) are expected to be some form of partial or full centralization involving functional splits.
In networks in which RAN functions are physically separated from one another, the transport network is the glue that holds the RAN together. It includes fronthaul (connecting the RU to the DU), midhaul (connecting the DU to the CU) and backhaul (connecting the CU to the mobile core) segments. Where RAN interoperability is needed, so is standardized transport.
In another trend, the concepts of open RAN are coming to the mobile transport network elements themselves — including disaggregation of the network operating system (NOS) and the underlying switching hardware. In the white box model, the disaggregated switching hardware and software are built to open requirements defined in groups such as OCP and TIP. The disaggregated cell site gateway (DCSG) segment for mobile networks, promoted by TIP, is one prominent example of the hardware and software disaggregation model. But the NOS/hardware separation model extends to other routing segments as well, including edge and core routing.
In the survey, Heavy Reading asked operators to choose the most important criteria in selecting a NOS supplier for disaggregated routing. Topping the list (and statistically tied) are the ability of the vendor to provide customer service for both NOS software and hardware (selected by 45% of respondents) and the maturity of the NOS software (selected by 44%). Third on the priority list is completeness of feature sets (selected by 40%).
These criteria align closely with operators' top disaggregation challenges (discussed in the previous blog) — specifically, technology maturity/available features and the internal process and operational models. If a vendor provides support for the full system, has a strong deployment record and offers crucial features, that vendor (or partnership of vendors) mitigates the primary challenges standing in the way of commercial deployments.
Although it is up to the vendors to meet the mark, operators have laid out a clear blueprint for overcoming their challenges.
Momentum around open and disaggregated networks is accelerating, particularly for the "partially" or "horizontally" disaggregated flavor in which the transponder vendor is chosen separately from the line system vendor. While components are supplied by separate vendors, the optical network is controlled and managed as a unified system (through open APIs).
Of the operators surveyed, 48% expect to multi-vendor optical networks in the 2021-22 timeframe, in addition to the 18% that have already deployed these networks is some form. Only 18% have no plans for open optical networks. Not surprisingly, North American operators are the most aggressive in their current deployments and timelines, but the trend is a global one.
The commercial adoption of pluggable 400Gbps optics will add fuel to the open optical trend over the next three years. It will be driven by the introduction of the OIF's 400ZR standard, as well as longer distance pluggable 400G variants typically identified as "400ZR+" or "OpenZR+" (though they are not part of the OIF work). The combination of pluggable, high bit rate optics, open line systems and multi-vendor management and control is set to drive a new era of open optics unlike anything witnessed in the past.
Looking for more information?
Check out the following report and webinar:
— Sterling Perrin, Senior Principal Analyst – Optical Networking & Transport, Heavy Reading
This blog is sponsored by Infinera, IP Infusion, Volta, and Telecom Infra Project.