ServersHardware Today: Caring for Your Cables

Hardware Today: Caring for Your Cables

ServerWatch content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.




Cable management is like the oft-ignored stepchild of server room management — everybody has loads of cables but finding the time to pay them the attention they deserve is another matter. If you don’t want that tangle of wires to come back and trip you up, now might be the time to bring some order into the chaos.

Cable management may seem mundane, but it is a vital for a data center to run smoothly. We offer some tips to bring order to the chaos.

According to Brad Wittenkeller, global marketing manager at wiring and communications vendor Panduit, a wide range of errors are being made in data centers around the country.

“Some of the biggest mistakes include poor planning for upgrading/scalability, and disregard for bend radius control of cabling and the impact on network performance,” said Wittenkeller. “Consideration also needs to be given to cable routing in high port-density applications, to the amount of space consumed by cabling, and to its impact on air flow.”

Cable management begins with racks and cabinets, which should provide ample vertical and horizontal cable management. Far from being merely aesthetically pleasing, well-organized cabling helps keep equipment cool by removing obstacles to air movement.

“Power and data cables can obstruct airflow causing damage to IT equipment that isn’t properly cooled,” said Russell Senesac, director of InfraStruXure Systems at power and cooling specialist APC.

The formula for Category 6 UTP cable, the most widely used type of network cabling, is to multiply the number of cables by the cable diameter (0.0625 square inches) by 1.30 (to prevent the space for cabling from being more than 70 percent full).

There is an equation that can be used to ensure that any rack or cabinet provides adequate cable management capacity. The formula for Category 6 unshielded twisted pair (UTP) cable, the most widely used type of network cabling, is to multiply the number of cables by the cable diameter (0.0625 square inches) by 1.30 (to prevent the space for cabling from being more than 70 percent full). For 350 cables, that equates to 28.44 square inches for the cable manager.

“Too often, companies place empty racks close together [and] give no thought at all to cabling and then quickly run out of space,” said John Schmidt, manager of Ethernet infrastructure at ADC, a vendor specializing in cable management and data center layout. “It is advisable to plan upfront so you have enough room for all the cabling you need at each side of the rack.”

ADC recommends cabling systems use common rack frames to simplify rack assembly and provide unified cable management. The company also recommends installing ample vertical and horizontal cable management within and between rack frames. The company further believes it is important to install both overhead and under-floor cable pathways to provide for orderly growth — under-floor pathways for permanent cabling and overhead for temporary cabling.

Not all vendors agree on best practices for the data center, however. Senasac of APC isn’t keen on under-floor cabling. The company prefers data and power cables be routed above the rack for easy identification and servicability.

A good model for data center layout, which includes some of the basics of cable management noted above, is covered in standards developed by the Telecommunications Industry Association (TIA). TIA 942 is a standard for network cabling infrastructure in data centers, while TIA 606 identifies key network components.

“TIA 606-compliant labeling improves the management and traceability and reduces total cost of ownership,” said Wittenkeller. “And don’t forget about proper grounding of racks and active equipment.”

Another tip concerns the best way to connect cabling. There are three methods of connecting equipment in the data center: direct connect, interconnect, and cross-connect. Many IT departments get lazy and stick with what appears to be the simplest option — direct connection. However, this is not a wise choice, as it means operators must hunt for cables with every change and pull them to the new location. With interconnection via a patch panel, the IS department need only reroute end cables when changes are made. While this is far more efficient than the direct connection, it is not as easy or reliable as the cross-connection method.

A centralized cross-connect patching system gives all network elements permanent equipment cable connections that are terminated once and never handled again. All changes are made using semi-permanent patch cord connections on the front of the cross-connect system, such as the ADC Ethernet Distribution Frame. This approach reduces the time it takes for adding cards, moving circuits, upgrading software, and performing maintenance. Instead of carrying out moves, additions, and changes on the backplanes of sensitive routing and switching equipment, they are done on the patching field. This lowers costs, speeds up network changes, and improves reliability.

>> A Fiber-Rich Diet

Get the Free Newsletter!

Subscribe to Daily Tech Insider for top news, trends & analysis

Latest Posts

Related Stories