Sarah Renée Ruepp

Green-aware routing in GMPLS networks

The increasing amount of traffic in the Internet has been accommodated by the exponential growth of bandwidth provided by the optical networks technologies. However, such a growth has been also accompanied by an increase in the energy consumption and the concomitant green house gases (GHG) emissions. Despite the efforts for improving energy efficiency in silicon technologies and network designs, the large energy consumption still poses challenges for the future development of Internet. In this paper, we propose an extension of the Open Shortest Path First - Traffic Engineering (OSPF-TE) protocol and a green-aware routing and wavelength assignment (RWA) algorithm for minimizing the GHG emissions by routing connection requests through green network elements (NE). The network behavior and the performance of the algorithm are analyzed through simulations under different scenarios, and results show that it is possible to reduce GHGs emissions at the expense of an increase in the path length, and, in some cases, in the blocking probability. The trade-off between emissions and performance is studied. To the authors knowledge, this is the first work that provides a detailed study of a green-aware OSPF protocol.

Label preference schemes in GMPLS controlled networks

The GMPLS assumption that all available labels are equal is reasonable in electronic networks but not always true in WDM optical networks where labels correspond to physical wavelengths. In this paper we present two schemes for collecting the preference for specific labels during GMPLS signaling. For this purpose a new use of the Suggested Label object is proposed, and a novel object called Suggested Vector is introduced. The approach is validated through simulations showing significant wavelength converter usage reduction in a WDM optical network

OSPF-TE extensions for green routing in optical networks

This paper proposes extensions to the OSPF-TE protocol to enable green routing in GMPLS-controlled optical networks. Simulation results show a remarkable reduction in CO2 emissions by preferring network elements powered by green energy sources in the connection routing.

Energy-aware routing optimization in dynamic GMPLS controlled optical networks

In this paper, routing optimizations based on energy sources are proposed in dynamic GMPLS controlled optical networks. The influences of re-routing and load balancing factors on the algorithm are evaluated, with a focus on different re-routing thresholds. Results from dynamic network simulations show that re-routing strategies can lower CO2 emissions compared to the basic energy source routing scheme, and a lower re-routing threshold achieves more savings. The increased blocking probability brought by using re-routing schemes can be compensated by applying load balancing criteria. A trade-off between blocking probability and obtained CO2 savings is studied.

Experimental Demonstration of Multidimensional Switching Nodes for All-Optical Data Center Networks

This paper reports on a novel ring-based data center architecture composed of multidimensional switching nodes. The nodes are interconnected with multicore fibers and can provide switching in three different physical, hierarchically overlaid dimensions (space, wavelength, and time). The proposed architecture allows for scaling in different dimensions while at the same time providing support for connections with different granularity. The ring topology reduces the number of different physical links required, leading to simplified cabling and easier link management, while optical bypass holds the prospect of low latency and low-power consumption. The performance of the multidimensional switching nodes has been investigated in an experimental demonstration comprising three network nodes connected with multicore fibers. Both high capacity wavelength connections and time-shared subwavelength connections have been established for connecting different nodes by switching indifferent physical dimensions. Error-free performance (BER <; 10-9) has been achieved for all the connections with various granularity in all the investigated switching scenarios. The scalability of the system has been studied by increasing the transmission capacity to 1 Tbit/s/core equivalent to 7 Tbit/s total throughput in a single seven-core multicore fiber. The error-free performance (BER <; 10-9) for all the connections confirms that the proposed architecture can meet the existing

Analysis of energy efficiency in dynamic optical networks employing solar energy sources

The paper presents energy efficient routing in dynamic optical networks, where solar energy sources are employed for the network nodes. Different parameters are evaluated, including the number of nodes that have access to solar energy sources, the different maximum solar output power, traffic type and the locations of solar powered nodes. Results show a maximum 39% savings in energy consumption with different increases in connection blocking probability.

A Framework for Label Preference in GMPLS Controlled Optical Networks

In GMPLS optical networks, the label (i.e., wavelength) selection process can be accomplished in a dynamic and distributed way by the RSVP-TE signalling protocol. In wavelength-continuous networks, the concurrent selection of the same label by distinct reservation attempts may lead to a connection blocking. In wavelength- convertible networks, the selection of a distinct label on each link may lead to a waste of wavelength converters, and thus to an increase of network cost and blocking probability. This paper presents label preference schemes that aim at improving the wavelength selection process during both provisioning and restoration phases. By exploiting the standard and novel RSVP-TE objects, reservation contentions are reduced in wavelength- continuous networks and converters are saved in wavelength-convertible networks.

Improved energy efficiency for optical transport networks by elastic forward error correction

In this paper we propose a scheme for reducing the energy consumption of optical links by means of adaptive forward error correction (FEC). The scheme worksby performing on the fly adjustments to the code rate of the FEC, adding extra parity bits to the data stream whenever extra capacity is available. We show that this additional parity information decreases the number of necessary decoding iterations and thus reduces the power consumption in iterative decoders during periods of low load. The code rate adjustments can be done on a frame-by-frame basis and thus make it possible to manipulate the balance between effective data rate and FEC coding gain without any disruption to the live traffic. As a consequence, these automatic adjustments can be performed very often based on the current traffic demand and bit error rate performance of the links through the network. The FEC scheme itself is designed to work as a transparent add-on to transceivers running the optical transport network (OTN) protocol, adding an extra layer of elastic soft-decision FEC to the built-in hard-decision FEC implemented in OTN, while retaining interoperability with existing OTN equipment. In order to facilitate dynamic code rate adaptation, we propose a programmable encoder and decoder design approach, which can implement various codes depending on the desired code rate using the same basic circuitry. This design ensures optimal coding gain performance with a modest overhead for supporting multiple codes with minimal impact on the area and power requirements of the decoder.

Out-of-sequence preventative cell dispatching for multicast input-queued space-memory-memory Clos-network

This paper proposes two out-of-sequence (OOS) preventative cell dispatching algorithms for the multicast input-queued space-memory-memory (IQ-SMM) Clos-network switch architecture, i.e. the multicast flow-based DSRR (MF-DSRR) and the multicast flow-based round-robin (MFRR). Treating each cell independently, the desynchronized static round-robin (DSRR) cell dispatching scheme can evenly distribute cells to the central switching modules, however, its frequent change of the input switching module connection pattern causes a serious OOS problem to the IQ-SMM architecture. Therefore large reassembly buffers are required at the output ports and high reassembly delay can degrade the multicast services. MF-DSRR can reduce the OOS problem and leverages the principle of DSRR to obtain a low complexity, however it fails to eliminate the in-packet OOS where cells of the same packet are disordered. Using more resources compared to MF-DSRR, MFRR is able to eliminate the in-packet OOS, resulting in a significant reduction of reassembly buffer size and delay.

Wavelength-Converter Saving Span Restoration in GMPLS Controlled WDM Optical Networks

We present two label preference schemes to reduce wavelength-conversion during restoration path setup in GMPLS controlled optical networks exploiting span restoration. The amount of required wavelength-conversions can be reduced up to 34 percent.