alex_aldo - Fotolia
Recently, someone tweeted --once again -- the YouTube video of a brown bear getting blown off a power pole in California. Naturally, this got me wondering: Whatever happened to IEEE 1901, also known as Broadband over Power Line (BPL)? Despite the awful arc flash, headlong fall and what I can only assume was a pervading smell of ozone and burnt fur, the bear survived. However, the same can't be said for Broadband over Power Line.
The Internet of Things (IoT) will need every trick in the last-mile book to connect the predicted 32 billion devices, and one might think BPL could supply enough niche connectivity to survive as one of those last-mile options, with at least small deployments for specific circumstances. IEEE 1901 has become remarkably quick and reliable in highly localized environments, so why hasn't it made the hop over the transformer and out to Internet connection points?
Everyday engineer is low voltage
Your everyday geek is low voltage. Yes, we'll play with mains power on occasion, but prefer to leave actual high-voltage engineering to someone else. Sure, electrical engineers do power-brick design all day, but this is something different. I'm talking about the narrow, hair-breadth space between the pedestrian -- though more easily employable -- engineering most of us do and the pyrotechnic engineering done by crews at places like Sandia (you know, the engineering that involves mounting locomotives on rocket sleds and crashing them into nuclear waste containers). Power systems engineering falls somewhere between the two types, and gurus of three-phase and high-voltage, direct-current engineering don't spend a lot of time working on frequency bridges and modems.
It's that challenge --the need to install, maintain and troubleshoot a box sitting high on a pole, safely bridging primary mains to secondary lines while carrying data past the distribution transformer --that kills BPL. We've all seen the bear on the pole video and we've all watched linemen with rubber boots, thick gloves and long fiberglass poles. Can you see yourself troubleshooting a network device up there? Nope, no thanks. Fatal arc flash has never been reported with 802.3at at 57 volts.
Cost the reason behind BPL's demise
The ultimate explanation behind the demise of Broadband over Power Line is actually cost rather than geeks' resistance to electrocution. Likewise, IoT's expansion will ultimately come down to cost. Reliable BPL above 50 Mbps is expensive to do well. BPL broadband providers that send expensive line crews out to fiddle with high-voltage data infrastructure can't compete with existing telcos that can provide 200 Mbps connectivity or more over existing low-voltage coax and twisted pair.
Additionally, as network administrators we have enough issues to deal with already in Layer 1-7 without adding serious RF interference into the mix. Yes, there can be some Layer 0 issues with signal-to-noise ratios even under a desk, but configuration and user demand overwhelmingly drive our help desk tickets.
The other challenge for power line companies can, however, affect any IT team. Power companies aren't generally focused on application delivery. It's not their fault -- their focus is energizing a power grid and worrying about environmental regulations -- but if reliable Layer 1 is already a problem for an organization, how can they be expected to have time to achieve great Layer 7 (application) user experiences? Our environments, however, offer no such excuse, and any network administrator should be thinking of applications and user experience first. When IT administrators fail at this, it's not that they don’t care or don't know that endpoint performance of apps is critical. Far from it, actually. In fact, slow or "broken" applications drive more tickets than anything else.
Better instrumented networks will win
First, remember that although we have our own issues with Layer 2-6 in the enterprise, it could be way worse. We could be one of the poor schlubs at the cable company breaking the bad news on a support call that there's nothing he can do to fix a signal power issue other than schedule a technician's visit. I can imagine them looking at the S/N chart for the modem, pausing apprehensively and knowing deep down that the yelling is about to start. For us, we instrument and poll everything possible with a variety of protocols, act on log events and traps, back up our configs daily and analyze flow data wherever we can get it. (You do back up your configs daily, right?) We have dashboards that reveal the most intimate nuance of our network's performance and generate reports for our managers, yet somehow find time to do actual planning and R&D.
So, the lesson of BPL is simple: New network technologies succeed when they are easy to manage by the experienced teams enterprises have in place today. If we can't easily manage it, we choose something else -- even if it's not as advanced. Technology is especially likely to succeed if it's compelling enough to encourage admins to stay abreast of accelerating technology. The ultimate success of IoT, cloud, SDN, you name it, will be determined by their ease of management and our commitment to do so.
Still though, it'd be funny to walk into a wiring closet and find a surprised co-worker squeezing a fingertip while the faintest hint of charcoal wafts in the air. "PoE RJ45 got my pinky!" Uh huh, sure it did. However, from experience I can say a POTS line, ringing 90VAC @ 20Hz, can indeed be a most unpleasant surprise between your thumb and index finger.
About the author:
Patrick Hubbard is a head geek and senior technical product marketing manager at SolarWinds. With 20 years of technical expertise and IT customer perspective, his networking management experience includes work with campus, data center, storage networks, VoIP and virtualization, with a focus on application and service delivery in both Fortune 500 companies and startups in high tech, transportation, financial services and telecom industries. He can be reached at Patrick.Hubbard@solarwinds.com.