Planning for success with cable management |Corning

by Cindy Ryborz, Marketing Manager Data Centres EMEA, Corning Optical Communications

There is a lot we can do to future-proof data centre infrastructures and prepare for unexpected incidents – a key component is cable management, which is essential for the performance of the network. Well-organised cabling means easier moves, adds and changes (MACs), bend-radius control and troubleshooting in case of issues, particularly compared to unstructured “spaghetti” cabling.

Today’s data centre networks are faced with constantly growing and changing demands. What seems like more than enough capacity today, will most likely not be sufficient for tomorrow.  COVID-19 has of course provided a prime example of how suddenly and unexpectedly our networks can be tested and how data centres need to scale.

To give an example, DE-CIX, one of the world’s leading internet exchanges, saw a record traffic peak of 9.1Tbit/s in March 2020 representing an increase of 800 Gbit/s. To put this into perspective, imagine the simultaneous transmission of up to 200 million HD videos. During the US presidential elections in November 2020, the all-time peak rose even higher, to 10Tbit/s.

Structured cabling helps to organise and manage cables, allowing better airflow and preventing loose cabling from tangling up, kinking, bending or even breaking fibres, all of which has an impact on the network performance.

Crowded cable trays, pathways or ducts as well as clusters of loose hanging patch cords in densely populated server racks may also create “hot spots” impacting the performance of network components or having them fail, creating possible delays or downtime – something we of course want to avoid.

The deployment of structured cabling, of course, is not a new concept and data centre environments continue to move away from the practice of patching and connecting equipment from point to point or as it is installed. A more recent development, which is impacting approaches to structured cabling in large projects is the growth of massive data centres beyond the confines of a single building or data hall.

These campus environments, with multiple buildings, require a high-fibre count cabling infrastructure, which can be a challenge to deploy.

There’s a number of solutions available to meet such high-fibre-count requirements. While splicing may be sufficient for ribbon cables of up to 864 fibres, when connectivity requirements are pushing fibre counts to 3456 and even 6912 fibres, pre-terminated trunks or pigtails with multifibre connectors are a much better choice.

The multi-fibre push-on (MPO) format dramatically reduces the amount of time, effort, and space required to install and deploy network technologies, particularly in parallel optic applications. So whether working with duplex, 8, or 16-fibre transmissions, the MPO connector scales to whatever technology is used – including parallel optic applications such as 400 Gb Ethernet - and allows for port breakout applications reducing the TCO (total cost of ownership) of a data centre deployment.

We worked with a large connectivity provider who needed an infrastructure upgrade. They had installed numerous duplex zipcords into a basket and because of the large number of connections (and thus cables), there was concern that some of those connections would be compromised and overall reliability would be impacted.

The solution, which encompassed high-fibre-count optical cables, connectors, and hardware hinged upon bend-insensitive fibres and ultimately ensured that the network could withstand the MACs that would occur in their MDA without disruption or downtime.

Another project, with a managed service provider to maximise capacity at its 9,717 square metre colocation data centre, required a new cable management solution to support its growth. Fundamental to the project was ensuring a high-density infrastructure solution by maximising port capacity within a small footprint and ensuring the infrastructure was flexible and scalable to changing business requirements and future growth.

The approach here was to provide a scalable fibre management solution for cross-connect applications in the data centre’s central hub. We used a frame design which provided optimised routing paths for patch cords, reducing the risk of entanglement, while the operations staff could install or remove a single patch cord in less than two minutes regardless of the cable route. To further simplify deployment and stock levels, the frame only required a single-length patch cord of just 4m to connect any port to any other port within the frame. <Download the full case study>

In some scenarios, disorganised cables can pose a genuine safety risk too. We worked with a multi-tenant data centre looking for a better way to organise their network. In the MDA, many of the patch cords were longer than the required distances, which was not only an eyesore, but also had safety implications. MACs were also difficult, requiring too much time to identify connections and increasing the risk of accidental disconnection.

In addition to the previously mentioned frame design that allows for growth and expansion of the network one frame, module, or fibre termination at a time, we also worked to organise their patch cords. A self-tracking system enabled simplified inventory management and by ordering the specific jumper lengths needed, the company was able to decrease clutter and clean up its interconnection room.

As data centres scale and become more complex, the demands for cable management complexities do increase but the solutions are already in place to meet them. Ultimately, not only is a well-planned cabling beneficial to the MACs of current customers, but organised cabling also attracts prospective end-users. The ability to confidently give customer tours is a major bonus, particularly for multi-tenant data centre owners who want to find a way to differentiate themselves and could be the deciding factor between a multi-year contract and lost business.