Dear Readers: This month I am passing off my responsibilities as author of this column to the industry-respected Vernon May, an expert in copper, DSL, and fiber. He has created, implemented and taught telecommunications technology on 6 continents for over 35 years, and has leveraged his service provider and vendor experience into that of trainer/coach, inventor, systems designer, program manager, product manager, and plant operations expert. He currently holds 2 US Patents, including #8,027,807 B2 on Remote DSL Circuit Evaluation.
My Favorite Spectator Sport Is Back in Season
By Vernon May
No, I don’t mean those large men that run around in a stadium on Sunday afternoons — I mean the fights between the DSLAM vendors and the Customer Modem vendors.
The disputes begin with a warm-up. A DSLAM/modem combination demonstrates interoperability issues in the field. Therefore, the local technicians are blamed immediately. The DSLAM vendor and modem vendor will agree on this point, so it is just a formality. This phase can last as long as 3 weeks. Meanwhile, subscribers are getting angry. Ultimately, the technicians are proven to be innocent.
Now the fight between the DSLAM and modem vendors begins in earnest. The finger pointing can become quite brutal at times. Blaming the other vendor first will generally score high, but "conclusive lab results" can end the game early.
Despite the high entertainment value, interoperability issues between the near end and the far end of a DSL circuit cause subscriber problems. With every step forward in DSL, the handshake between the DSLAM and modem becomes more complicated, and interoperability issues are inevitable. They must be addressed objectively, quickly, and effectively. However, we learned this lesson the hard way.
G.992.1 and G.992.2- ADSL(1)
Different technologies were fighting for dominance. If any modem and any DSLAM combination synchronized at all, it was cause for celebration. Universal interoperability was only a dream.
In order to have any customers with working DSL service, the industry was forced to have large interoperability events where the DSLAM and modem vendors worked together.
G.992.4 and G.992.5- ADSL2/2+
ADSL2 was never really deployed alone, and was melded into ADSL2+. G.DMT had won dominance, and with most developers at least trying to follow the specs, interoperability became an achievable goal. Despite this, interoperability issues between the DSLAM and modems were and still are common.
G.993.1 and G.993.2- VDSL1 and VDSL2
VDSL introduced spectral profile options and multiple upstream/downstream bands. The specifications were good, but nobody could make them work in the beginning. Therefore, VDSL was initially deployed
as "VDSL 1.5", which was not really a spec; it was a marketing name for not following the spec. Broadcom (https://www.broadcom.com/) finally released true VDSL1/2 chipsets, so interoperability again became achievable. Still again, there were and are issues.
The best first step in resolving interoperability issues between DSLAM vendors and modem vendors is to give the local technicians some credibility from the beginning. Field and lab conditions are quite different. Sure, the technicians are wrong sometimes, but they always offer important insight into the subscriber experience.
The key is to maintain the common goal of providing the most reliable service possible, without regard to who is to be blamed. Sometimes, a modem issue is more easily fixed in the DSLAM. The concept of every issue being "our issue" serves us well.
The interoperability battle for vectoring in North America is beginning. The winners here should have the advantage in the G.fast battle waiting on the horizon. It looks to be another exciting season.
Vernon and I would love to hear your comments on how you are coping with the issues around DSL implementation. Contact me at [email protected] or call 831.818.3930.