Preparing for PoE Plus and Energy Efficient Ethernet

Power over Ethernet has been around for some time, but the standard is evolving and this could have implications for designing corporate networks, writes Carrie Higbie, president of the Blade Systems Alliance.

Power over Ethernet has been around for some time, but the standard is evolving and this could have implications for designing corporate networks, writes Carrie Higbie, president of the Blade Systems Alliance.

Power over Ethernet describes a system to transmit electrical power, along with data, to remote devices over cables in an Ethernet network. This technology is used in powering IP telephones and wireless Lan access points.

Standards body the IEEE is working on Power over Ethernet Plus (PoE Plus), which will effectively double the amount of power provided over a data cable. This will enable a wide range of capabilities, such as PTZ (pan tilt and zoom) for video cameras, extended range wireless access points, and extended video capabilities within VoIP phones.

PoE Plus is geared towards providing greater power (>30 Watts) for hardware that requires more than 12.95W at the end device. Heat dissipation is directly affected by conductor size and resistance and it is likely that newer 10gbps copper cabling systems, such as category 7/class F and category 6A/class EA are likely to prove to be better choices for higher power due to larger conductor size.

At present, category 7/class F is also the only published standard to support 10GBase-T over 100 metres. Although the new standards generally try to address installed cabling plants, the safest bet for new installations is to use the highest performing cabling possible as there may be less restrictions in the future.

Facilities may need to install higher amperage breakers and, in some instances, move to higher gauge electrical wire to support the increased power demands in the telecoms areas. Whatever type of power sourcing unit you install, it is important that it be able to throttle down power when it is not needed.

Other benefits that may be realised over time revolve around the Energy Efficient Ethernet work being done by the IEEE. One disadvantage to powered patch panels is that the systems are installed into an infrastructure that is designed to last 10 to 15 years. This may limit those installations from taking advantage of the new energy efficient technologies being developed.

Ethernet ports draw power when idle. With green initiatives on the rise, work on Energy Efficient Ethernet aims to find ways to shut down the ports during idle states, throttle back bandwidth, and may even translate into management schemes that would allow VoIP phones, for instance, to power down to low state power after hours.

In a switched environment, power can be dedicated on the fly, rather than requiring a patch cord change or retermination. The newer switches also allow for static power allocation as well as dynamic. Network managers can dictate a maximum amount of power that will be supplied to a port, although the device may not draw that amount of power at all times, the remaining power can be returned to the power budget when the device is in a quiescent state.

The ability to upgrade to more efficient power as technologies evolve without having to reterminate connections will be labour savings. Likewise, a single point of management and single connections into redundant power as opposed to two for mid-span may mean the difference between multiple uninterruptible power supply units as opposed to single ones. Mid-span devices outside of patch panels also take up additional rack space which is not available or a challenge in some environments.

Read more on Network hardware