JOURNAL OF LIGHTWAVE TECHNOLOGY, VOL. XX, NO. X, OCTOBER 2015 1 Experimental End-to-End Demonstration of Shared N:M Dual Homed Protection in SDN-controlled Long Reach PON and Pan-European Core S´ eamas McGettrick, Frank Slyne, Nattapong Kitsuwan, David B. Payne, and Marco Ruffini, Senior Member, IEEE Abstract—Seeking reduction of capital and operational costs on next generation fibre networks is the holy grail of network planning and deployment for all operators world wide. In particular, efficient deployment strategies of next generation fibre access networks is of paramount importance to enable wide scale ubiquitous deployment of high-speed broadband services. Exploiting the large capacity of fibre networks to support heterogeneous services from residential and business users is a promising strategy towards this goal. Long-Reach Passive Optical Networks (LR-PON) is one such strategy which also adopts greater sharing of active and passive components, together with consolidation of central offices, to reduce capital and operational expenditures. However due to its long reach and large split ratio, protection mechanisms become a major consideration when designing an LR-PON, as a single feeder cable cut could disrupt services for several thousand users. In this paper we demonstrate fast restoration of LR-PON services using a dual-homed, shared- OLT protection mechanism. Our end-to-end testbed connects our optical access broadband laboratory operating on custom built LR-PON ONU and OLT FPGA prototypes with a Europe-wide testbed core network (G ´ EANT). We use an SDN control plane to manage the dual-homed N:M protection switching and traffic reroute in the core, and achieve access protection and end-to-end services restoration times of 40 ms and 80 ms respectively. Index Terms—Long-Reach Passive Optical Network, GPON, XG-PON, N:M protection, SDN, end-to-end, service restoration, converged architecture. I. I NTRODUCTION M ANY operators worldwide are currently upgrading their access network to fibre network, and there is a strong push on developing efficient network designs able to reduce capital and operational costs for the deployment of the current and next generation fibre access networks, while increasing the revenue they can generate. Firstly, a fiber access network with its high capacity lends itself to carry much more than residential traffic. This could enable the access network to be used in new ways, for example to offer dynamic on demand services to residential and small/medium businesses alike, and to be utilized for mobile X-hauling (i.e., including backhaul- ing, fronthauling and any other means of transporting mobile This material is based upon works jointly supported by the Science Foun- dation Ireland under Grant No. 10/CE/I1853 and European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement no. 318137 (Collaborative project ”DISCUS”). S. McGettrick, F. Slyne,N. Kituswan, D.B. Payne and M. Ruffini are with CTVR/CONNECT, The Telecommunications Centre, Trinity College Dublin, Ireland e-mail: (mcgettrs@tcd.ie). Manuscript submitted Oct, 2015 access data). Such a heterogeneous utilization of the fiber access network will increase its efficiency of use, increasing the revenues it can bring and thus the time to positive cash flow. Secondly, more efficient network architectures can be investigated, aimed at reducing the initial deployment costs as well as its future upgrade and maintenance requirements. One such access architecture is the Long Reach Passive Optical Networks (LR-PON), which introduces larger split ratios and longer reach compared with current PON systems, allowing the bypass of a portion of the metro transmission network [1]. A significant aspect of network design is that of providing protection mechanisms, which are essential to ensure ade- quate network availability. However, providing redundancy adds costs to the network and so protection mechanisms are mainly limited to the core and metro networks, where the cost of adding redundancy to the network can be shared among many users. Access networks for residential users are typically unprotected, and this is reflected in the terms of Service Level Agreements (SLAs) which generally allow for service restoration times of days or weeks. Access networks for business users can instead operate on much stricter SLAs, although businesses normally apply for private line connec- tions and incur high cost to guarantee protection. While ultra- high availability can still be guaranteed to customer willing to absorb its cost through bespoke solutions, a multi-service access network as the one we envisaged requires offering adequate protection at acceptable price in order to satisfy a wide variety of services and customers. This is especially true for architectures like LR-PON that provide cost savings by promoting larger customer aggregation and bypass of transmission networks that are normally pro- tected: thus special consideration needs to be taken to ensure appropriate protection strategies. Figure 1 compares the layout of a future LR-PON to that of current PON architectures. Similarly, Table I summarizes the points of failure for the LR-PON and current PON systems (XG-PON in this case) together with an approximation of the number of people affected by such a failure. We can see that since the LR-PON also operates as a metro aggregation point, the likelihood of one individual failure affecting a large number of customers is higher compared to a conventional access network: for this reason protection mechanisms become a requirement in LR-PON. A dual homing architecture is also important as it provides additional resilience through link and node diversity, and it can be implemented as far as the constraint over the