During the 25th Next Generation Optical Network Forum (NGON), Italo Busi, a senior expert from Huawei European Research Center, noted that the IPoWDM solution (i.e., a solution with colored pluggable modules on the routers) is still unable to address four challenges in the 400G era: poor transport performance, complex O&M, difficult network evolution, and high network TCO. In this context, the IP+WDM solution (i.e., a solution with gray optical modules on the routers and WDM interfaces on the optical devices) has obvious advantages, and so remains operator's best option for building networks.
With the explosive growth of 5G, FTTH, and DC services, network traffic surges. According to Omdia, the annual growth rate of network traffic will exceed 30% in the next five years, posing higher requirements on the bandwidth and overall cost of IP and optical transport networks. Currently, the router interface rate has increased from 100GE to 400GE, and the optical transport network has evolved to single-wavelength 400G and 800G. The IP+WDM solution enables flexible scheduling at the IP layer, as well as stable and reliable optical transport, satisfying the development and evolution requirements of all-scenario network services. Several device vendors advocate the use of 400G ZR+ IPoWDM solutions to reduce network costs. However, according to comparative analysis results, the IPoWDM solution still faces several challenges and problems to become the ideal solution for operators.
- Challenge 1: Poor transport performance & limited application scenarios. 400G ZR+ optical modules are based on standardized service modulation and FEC algorithms. They have much lower performance than traditional 400G colored optical modules. 400G ZR+ optical modules generally achieve a transmission distance of less than 200 km on live networks, making them suitable for only a few short-haul scenarios. In addition, 400G ZR+ optical modules consume 100% more power than traditional 400GE gray optical modules, and therefore services cannot be fully configured on the service board ports of IP devices, leading to about 50% ports unavailable and wasted.
- Challenge 2: Complex O&M. IP and optical transport are different industry systems. They have independent complex technologies in terms of design, configuration, and commissioning principles, and also have their own O&M functions. As such, it is difficult for O&M engineers to master both systems. In the IPoWDM solution, technology and administrative boundaries are blurred, so it is difficult to locate and analyze faults, especially irregular fault events. For example, it is difficult to determine whether a fault is caused by the IP protocol or OCh/OMS performance at the optical layer. This also leads to the fault demarcation is unclear; even if the colored router and L0 ROADM optical layer are provided by the same vendor, it is difficult to demarcate or locate faults for them.
- Challenge 3: Difficult network evolution. Currently, the IPoWDM solution supports only 400G and mature C96 fiber spectrum. There is no clear plan or roadmap for evolution to 800G or 1.6T nor to support wider optical spectrum (including super-C+L+S band). As a result, the single-fiber capacity will be limited to 25.6T for the foreseeable future.
- Challenge 4: High TCO. The IPoWDM solution was initially developed to reduce the TCO of networks. On live networks, the port rate of IP routers and WDM devices must match the actual traffic and transmission distance to reduce the per-bit cost. Now, the IPoWDM solution provides only one type of fixed 400G interface, and so cannot meet various service rate and transmission distance requirements on live networks. In most scenarios, the IPoWDM solution has high TCO. In addition, IPoWDM typically uses a hop-by-hop forwarding network architecture: as such, services need to undergo forwarding and optical-electrical conversion multiple times, resulting in high network latency and costs. The IP+WDM solution can match customers' service rates and implement one-hop optical-layer connection over longer transmission distances, thereby reducing service latency and costs.
To sum up, the IPoWDM solution cannot eliminate the four major challenges and problems: poor transport performance, complex O&M, difficult network evolution, and high TCO. In the 400G era, although the ZR+ technology innovation makes the IPoWDM solution applicable to some metro short-haul DCI scenarios, the IPoWDM solution still has disadvantages in most other scenarios. For long-term development of carrier networks, the IP+WDM solution remains the mainstream.
