“People are familiar with the constant change of the data center environment, but some changes are relatively far-reaching and disruptive in the long-term. Precisely, it is not just data centers (including hyperscale, global scale, multi-tenant, and enterprise data centers) that are affected by this fundamental change, but in the ecosystem, from designers, integrators and installers to OEMs and infrastructure All partners need to adapt to this.

“

Wu Jian, Technical Director of CommScope North Asia

People are familiar with the constant change of the data center environment, but some changes are relatively far-reaching and disruptive in the long-term. Precisely, it is not just data centers (including hyperscale, global scale, multi-tenant, and enterprise data centers) that are affected by this fundamental change, but in the ecosystem, from designers, integrators and installers to OEMs and infrastructure All partners need to adapt to this.

At present, we are witnessing a new round of speed upgrades. Large operators are now transforming to 400G applications and are planning to upgrade to 800G. So why is this latest leap so significant? First, the upgrade to 400G, 800G, and eventually 1.6T and 3.2T officially marks the beginning of the octal era, which will bring some fundamental changes that will affect everyone.

What factors are driving changes in data center infrastructure?

Within data centers, the global increase in data consumption and resource-intensive applications such as Big Data, Internet of Things (IoT), Artificial Intelligence (AI), and Machine Learning (ML) is driving the need for higher capacity and lower latency . At the switch level, faster, higher-capacity application-specific integrated circuits (ASICs) make this possible. Data center managers are faced with the challenge of how to configure more ports with higher data rates and higher optical paths, which requires more flexible deployment options to achieve complete expansion. Of course, all of this is taking place in the context of a new reality in which data centers have to do more with less physical and financial resources.

While it is ultimately up to the data center network manager to ensure that its own infrastructure gets the job done, its partners (including installers, integrators, system designers and OEMs) all play an important role. The value of physical layer infrastructure is largely determined by how easy it is to deploy, reconfigure, manage and scale.

Clear standards for future-proof flexible fiber optic platforms

Several years ago, shortly after launching a high-speed network upgrade platform, CommScope started looking at the next generation of fiber optic platforms. We asked our customers and partners, “Based on your current understanding of network design, migration, and installation challenges and application requirements, how would you design your next-generation fiber platform?” The answers were unanimous—easier and more efficient Upgrade to higher speeds, ultra-low-loss optical performance, faster deployment, and more flexible design options.

Combining these comments, and based on CommScope’s own more than 40 years of network design experience, we have identified several key design requirements to address changes affecting our data center customers and their design, installation and integration partners.

Application-Based Building Blocks

In general, application support is limited by the maximum number of I/O ports on the front of the switch. For a 1RU switch, the capacity is currently limited to 32 QSFP/QSFP-DD/OSFP ports. The key to maximizing port efficiency is maximizing the use of switch capacity.

The traditional four-channel four design provides a stable upgrade to 50G, 100G and 200G. But at 400G and beyond, the 12- and 24-fiber configurations used to support four-way applications are less efficient, leaving a lot of capacity stuck on switch ports, which is where octal technology comes into play.

Beginning with 400G, Octal Octal technology and 16-fiber MPO breakout are the most efficient multi-pair building blocks for backbone applications. The move from a four-way deployment to an eight-way configuration doubles the number of branches, thereby freeing network managers from the need for certain switch layers. Additionally, today’s applications are designed for 16-fiber cabling. Therefore, supporting 400G and higher applications with 16-fiber technology enables data centers to maximize switch capacity. This 16-fiber fiber design includes matched transceivers, backbone/array cables, and distribution modules and can be a common building block for data centers going from 400G to 800G, 1.6T and beyond.

However, not every data center is ready to say goodbye to traditional 12- and 24-fiber deployments, and they must also be able to support and manage applications without wasting fiber or losing port count. Therefore, efficient application building blocks for 8-fiber, 12-fiber and 24-fiber configurations are also required.

Design flexibility

Another key requirement is a more flexible design that enables data center managers and their design partners to quickly reallocate fiber capacity across the patch Panel and adjust their network to support changes in resource allocation. One of the ways to achieve this is to develop built-in modularity of panel components, enabling consistency between the point of delivery (POD) and the network design architecture.

In traditional fiber platform designs, components such as modules, termination boxes, and adapter packs are panel-specific. Therefore, changing a component with a different configuration also requires replacing the panel. The most obvious impact of this limitation is the additional time and cost involved in deploying new components and new panels. At the same time, data center customers also face additional product ordering and inventory costs.

In contrast, a design where all panel components are essentially interchangeable and can fit into a single universal panel will enable designers and installers to rapidly reconfigure and deploy fiber optic capacity in the shortest possible time and at the lowest cost. As such, it will also enable data center customers to streamline their infrastructure inventory and its associated costs.

Simplify and accelerate fiber deployment and management

A final key criterion identified by CommScope’s research and design efforts is the need to simplify and accelerate the routine work of deploying, upgrading and managing fiber infrastructure. While panel and blade bay designs have seen incremental improvements in function and design over the years, there is still plenty of room for improvement.

In addition, the issue of polarity management is also worth mentioning. As fiber deployments become more complex, it becomes more difficult to ensure that the transmit and receive paths are consistent throughout the link. In the worst case, ensuring polarity requires the installer to flip the module or cable assembly. Errors may not be discovered until the link is deployed, which can take additional time to resolve.

Brand new Propel^TM solution

Based on information collation, subsequent design and engineering work, CommScope has launched a new end-to-end high-speed modular fiber platform Propel solution, which we believe will significantly improve the long-term performance and value of today’s data center networks. The Propel platform is rigorously designed around key criteria such as application-based building blocks, design flexibility, and deployment speed.

The Propel solution is the first global fiber platform to incorporate native 16-fiber technology, supporting both 8-fiber, 12-fiber and 24-fiber applications. Therefore, it provides a single platform that can support multiple generations of networks. In addition, it is optimized for ultra-low loss optical performance. Precision, bevel-polished APC endfaces, and a proprietary system for precise fiber alignment ensure ULL signal transmission over longer distances.

All modules use enhanced B polarity uniformly, so there is no need to flip modules and components, and all connectors use consistent keys to ensure transmission to the correct receiver. Additionally, QR codes are incorporated into all panels and assemblies, enabling installers and technicians to verify optical performance and access various guides and product documentation from the phone.

At the component level, four different, fully interchangeable fiber module sizes simplify design and consistency between network applications and cabling options. All fiber optic modules fit into a single universal panel design, further simplifying upgrades and changes without field adjustments. The universal panel design also provides front and rear access, making modules, adapter kits and cable assemblies easier to install and service in tight spaces. Inside the panel, the blade bays are universal in design and accept any module type. There is also a unique design that allows the blade carrier to be removed and reinserted by a single technician when necessary. At the same time, cable management is made easier, and improved routing and safety clips can be easily snapped on and off the blade.

The Propel platform is a greener and more sustainable solution. It is able to support multiple upgrade cycles, reduce fiber optic component replacement and its associated carbon footprint, and enable data centers to take advantage of the increasing energy efficiency in each new generation of switches.

Collaborate to connect the present and the future

Data centers are rapidly evolving as data speeds and infrastructure complexity increase. This evolution is particularly evident in hyperscale environments, as channel speeds are accelerating to 400G, 800G and beyond, and the number of fibers at each layer of the network is multiplying. Managers and maintainers of data center infrastructure face the unprecedented challenge of making their networks future-proof and as efficient as possible. To do this, they needed a highly adaptable and robust high-speed fiber solution, as well as an infrastructure partner that could meet their global, always-on connectivity requirements.

By working closely with network managers, designers, integration specialists and installers, CommScope has identified the keys to the success of future network infrastructures. Based on more than four decades of network design experience and global engineering resources, CommScope develops and delivers fast, flexible and future-proof Propel solutions.