Archive

SWITCHED ENET: ONE WAY TO AUTOMATE
AND IMPLEMENT FIBER TO THE PREMISES

    By William Flanagan, Editor & Publisher

Last issue we defined the "everythings" that comprise the services ("Everything over IP!") on a converged IP network:  data (Internet access), voice, and video.  Talking as we were of IP, the Layer 3 protocol, we mostly ignored the infrastructure that carries IP.  Given sufficient bandwidth, adequate availability ("up time"), and not too much latency, most anything will do.

Applying the KISS principle, what technologies stand out as possibilities for the broadband local loop?  We know (and mostly love) DSL, Cable Modem, and satellite, each of which implies a specific link-layer protocol.  All of them also have some difficulty guaranteeing throughput to individual users--upstream, the customer often sees a choke point in the DSLAM, headend, or transponder.  Carriers have ways to make you happy, but generally don't bother as it is too expensive to configure and maintain.

That attitude was (perhaps) acceptable when the only service was fixed-price Internet access, and that only for web surfing, etc.  Voice and video, however, want some assurances on throughput, availability, and latency.  And aren't those the same qualities sought by gamers?  It isn't the sure of hand who necessarily win those combat games, it's the swift of access.  High-definition TV has its own requirements.

Just so happens that most of these problems (how to guarantee T/A/L) have been solved in routers and switches.  They have anticipated demand for Quality of Service (QoS) by years.  But how do you get those benefits into the local loop, far from the LAN?

On fiber, of course.  Now that the FCC has exempted new fiber deployments (at least for businesses and apartment buildings) the Local Exchange Carriers (LECs) can keep competitors out of new broadband outside plant, again enjoying their "natural monopoly" (at Layer 1, the physical layer in the protocol stack).  

There is still a question of what protocol to apply at Layer 2.  In addition to the ones now in use (ATM over DSL, etc.), several options based on Ethernet are reaching the market.  This issue we'll cover switched Ethernet, just like on a LAN.  Next time we'll look at Resilient Packet Ring (RPR).


Switched Enet

You know it from your wiring closet.  That stack of boxes (or large chassis with hundreds of inputs) linking desktops into the LAN backbone.  All modern switches run at 100 Mbit/s;  most also do either 10 or 1000 or all three, autosensing on the same card and port.  Switches are smart in other ways, too.  They can police traffic, control access, and guarantee throughput (reserve bandwidth).  That's just the ticket for two of the T/A/L requirements.

Standard Ethernet is as close to plug-and-play simplicity as we can hope for today.  It's protocols set up for any existing topology, then adapt to changes in cabling and equipment.  User plug in and get basic connectivity service automatically:  DHCP, bootp, Dynamic DNS, etc. greatly reduced administrative chores.  Carriers want to be slightly less than fully automatic--got to include access controls related to billing.

There have been some competitive LECs who tried to deliver Ethernet over fiber based on Gigabit Ethernet in the local loop.  They used commercial hardware, which worked (it had solved most of the problems in the LAN)--when it worked.  Standard routers, particularly the small ones sized for premises locations, aren't built for "five 9's" availability (5 min. of downtime per year, including maintenance).

The LECs wouldn't use equipment that can't live in pedestal cabinets, underground vaults, and other spaces that lack the environmental controls of the average office closet.  Can't blame them.  No point in deploying a converged infrastructure that lacks the "A" in T/A/L.

An example of the new solutions that cover all of T/A/L is the QoStream family of devices from Amedia Networks (www.amedianetworks.com):
--Father Bear is the CS1200, backplane capacity of 120 Gbit/s and a reach of 50 km on its fiber interfaces.  It's the "headend" feeding up to 50,000 subscribers.
--Mother Bear is the remote terminal, AS5000, 1-5 Gbit/s capacity that connects up to 48 subscribers on fiber (or copper if the distance is short enough).
--Baby Bear for the premises is the PG1000, the access device with POTS and Ethernet ports, internal battery backup, and a wireless access point.  A DSP digitizes the voice and provides other telephony features.

For LECs, the designs are "hardened" to survive in harsh surroundings, without air conditioning.  Designs and technology licensed from Lucent provide the basis for high availability (calculated so far;  they are too new to have extensive experience in the field with up time). 

The proposition Amedia stresses is the fact that each device is an Ethernet switch--with all the benefits appertaining thereto.  
-- Switching in hardware ensures low latency.  
-- With ever-less-expensive switching hardware, the functionality to differentiate classes of service becomes affordable for residential and small business customers.  
-- Switches can equitably divide the large capacity of fiber out of the central office and guarantee 100 Mbit/s to each premise.
-- "Smarts" in the switch can recognize local voice calls, minimizing traffic back to the CO.
-- Packet replication in the outside plant (for multicast video) reduces the number of streams for each program (perhaps to only one per distribution cable).

For privacy, Ethernet offers virtual LANs.  While not ultimately secure, vLANs under management of the carrier are good enough for most.    For better security, and some day for traffic engineering, look for MPLS switching to carry the Ethernet frames.

As equipment designed for carriers, this bunch includes the knobs and handles for remote management and fault isolation.  The boxes have SNMP.  The element manager (EM), called QoStream Director, offers CORBA on its northbound interface to the OSS.  This form of EM is important not only to blend new equipment into an order entry system, but also as a mechanism to support self-administration of services through a web portal.  Getting closer!


Next issue:  Resilient Packet Ring (RPR).

Write if you have a favorite access technology you'd like to see covered in ViewsLetter:  editor@viewsletter.com

============ + ===== + ===== + ===== + ============
PON  MARKET  REVIEW
      For a close look at more than two dozen PON companies, you are invited to purchase the report, "PON Industry Players--2004" available from Flanagan Consulting.  Offered on paper and a PDF file via email, this 25-page document describes each company's products, shows which market segments they participate in, and provides current contact information.  Either form is priced at US$60.00, payable by check to Flanagan Consulting, 45472 Holiday Dr. #3, Sterling, VA 20166.
============ + ===== + ===== + ===== + ============

FLANAGAN  CONSULTING 
-- Call us for a vendor-neutral network architecture and strategy for expansion or convergence.  We know voice AND data--and how to avoid expensive bear traps on the migration path, such as security arrangements.  
--Working on product positioning or a marketing message for telecom?   Yes, we've done that--for hardware products and carrier services.
-- Need an Expert Witness?  Associates at Flanagan Consulting have aided in many legal proceedings involving telecom intellectual property and technology.  
--For RFP preparation, bid analysis, proposal evaluation--call us.  We have current experience in Federal network procurement processes.

"We Have the Experience."

-- Special thanks for supporting ViewsLetter to www.webtorials.com, your best source for communications tutorials and white papers.
============ + ===== + ===== + ===== + ============