where operators connect to their customers,” the ONF says. “CORD is a project intent on transforming this edge into an agile service delivery platform enabling the
operator to deliver the best end-user experience along with innovative next-generation services.
“The CORD platform leverages SDN,
NFV and cloud technologies to build agile
data centers for the network edge.,” ONF
continued. “Integrating multiple open
source projects, CORD delivers a cloud-na-
tive, open, programmable, agile platform
for network operators to create innova-
tive services.”
CORD is packaged into three solu-
tions for different market-use cases, ONF
explained. M-CORD supports 5G mobile
edge services with disaggregated and vir-
tualized radio, and an open source mo-
bile core. R-CORD supports residential
subscribers over wireline access tech-
nologies like GPON, G.fast, 10GPON
and DOCSIS. E-CORD supports enter-
prise services such as virtual private net-
works and application optimization (soft-
ware-defined WAN) over metro and wide
area networks.
The VOLTHA 1.0 release is a notable
milestone for the CORD project. AT&T’s
Barker stated that major software releases
like it “are necessary to fulfill our vision
of a software-defined network, which
employs NFV. We expect to have 55 percent of our networks virtualized by the
end of 2017. We aim to have 75 percent of
our traffic on our software-defined network by 2020, and we’re pushing hard to
beat that goal.
“Open software efforts benefit the industry because we rely on the active participation and feedback form a large
community of developers,” he added.
“Developers can improve, add, and in-
fluence changes to the software that
will help us deliver XGS-PON technology
to customers quickly. We are currently
performing proof-of-concept testing of
VOLTHA in our labs and are planning
to deploy XGS-PON field trials before the
end of 2017.”
How POL fits
Back to the chat with the Tellabs sales
exec who mentioned passive optical
LAN and SDN in the same sentence … it
will be a very long time before anything
like VOLTHA makes its way into main-
stream enterprise networking. But SDN
is a timely topic for the LAN. In a docu-
ment aimed at federal-government us-
ers, Tellabs declares that passive optical
LAN offers the best architecture for soft-
ware-defined LANs. It explains that as
government network administrators eval-
uate the merits of SDN functionality in
buildings and across campuses, they are
doing so “under the assumption that SDN
fixes traditional LAN operational efficien-
cies, security and reliability shortcomings.
However, what they don’t realize is that by
bolting-on SDN as an overlay to a legacy
LAN design, they leave the inherent weak-
ness of traditional LANs.”
Pointing the finger at the traditional
LAN architecture, Tellabs further con-
tends, “Adding complexity with SDN can
marginally improve LAN operational effi-
ciencies, security and reliability, but by in-
troducing more sophistication, the fixes
can negatively contribute to the same at-
tributes they were intended to repair.
Furthermore, there are alternative means
of addressing the underlying fundamental
faults relative to traditional LAN … that
specifically fix root problems.”
Passive optical LAN, Tellabs explains,
is one such alternative means. The com-
pany points out the following potential
pitfalls of implementing SDN as an overlay
to a traditional LAN.
• Access, aggregation, distribution, and
work-group switches are complex,
full-functioning devices, representing
potential security weaknesses.
• Complex full-functioning switches
spread across buildings and a cam-
pus equals distributed intelligence and
management at each port, thereby re-
quiring local provisioning, trouble-
shooting and management of high-
er-level IP and Layer 3 functions
at each port.
• Adding SDN protocols to existing full-functioning switches inserts security, operation, and reliability complexities.
Conversely, Tellabs argues, an optical LAN “marries the best features of passive optical networking with advanced
Ethernet functionality. It does so within
the framework that matches cloud, wireless, hosted/managed services, data center and SDN architecture—all of which
have the common trait of having centralized intelligence and management.”
Plus, a passive optical LAN can define
network resources in software, and dynamically allocate them based on real-time demands.
Furthermore, the company stresses,
passive optical LAN facilitates SDN im-
plementation in part because “simple un-
managed ONTs [optical network termi-
nals] are better suited for SDN rather
than complex full-functioning traditional
switches,” and because a passive optical
LAN “will allow a mixture of G-PON, XGS-
PON, and NG-PON2 [40G] technology
choices simultaneously, without the rip-
and-replace of today’s infrastructure.”
As IBM’s white paper pointed out, op-
tics changed service-provider networks
in the 1980s. It was about 2010 when pas-
sive optical LAN technology took hold in
enterprise networks. It may be decades
before the fruits of the ONF’s efforts are
enjoyed by enterprise networks—if they
ever are. Nonetheless, proponents of pas-
sive optical LANs are pointing to history
to make their case for what the future
will hold. u
Patrick McLaughlin is our chief editor.
