Chankyun Lee

IP-Over-WDM Cross-Layer Design for Green Optical Networking With Energy Proportionality Consideration

Energy consumption in a network is becoming the physical limit against the everlasting increase of network traffic in the near future. Current network equipment consumes energy inefficiently due to poor energy-load proportionality of network equipment which causes a vast amount of energy consumption even in an idle state. We investigate and model the energy consumption of packet optical transport networks. Employing integer linear programming, we model and optimize the energy consumption of IP/WDM (IP/WDM) networks with traffic-grooming cross layer designs. Our study shows that energy savings in IP/WDM networks with traffic grooming is not effective unless IP layer equipment is energy proportional. In addition, different network planning schemes are required with respect to energy proportionality of network equipment.

Power and cost reduction by hybrid optical packet switching with shared memory buffering

The power consumption of a network becomes a critical performance metric for a petabit-per-second scale Internet, in addition to the system cost. We investigate the power and cost efficiency of all-optical and electrical technologies to understand the design principle for petabit-scale hybrid optical switch nodes for optical packet and burst switching. In order to achieve an efficient hybrid design, we point out the interesting observation that the use of long packets switched by passive-medium photonic switches can substantially reduce power consumption and system cost. For contention resolution, shared electronic memory buffers are more effective and power efficient than fiber delay line buffers. An empirical optimization exhibits that a traditional all-optical solution is not the best choice for power and cost efficiency. Using passive-medium photonic switches with electronic memory buffers, we can introduce a substantial savings in power and cost, approximately half of those of commercially available optical cross-connect systems, with a packet loss rate as low as 10-6.

Traffic grooming for IP-Over-WDM networks: Energy and delay perspectives

As energy consumption has become a major concern of networks, traffic grooming studies on Internet protocol over wavelength division multiplexing (IP-over-WDM) networks have been revisited for energy-efficient optical network applications. Traffic grooming techniques can save on the number of installed lightpaths and corresponding IP router line cards at the cost of traffic processing overhead. Energy awareness in networking device development has initiated improvement of the energy-traffic proportionality of networking devices. Accordingly, networks will consist of networking devices with a wide range of energy-traffic proportionalities. Therefore, previously known traffic grooming techniques may not achieve energy-efficient networking in high energy-traffic proportional networks. Moreover, the traffic grooming technique tends to prefer longer paths, which are responsible for longer transmission delays. These two concerns trigger investigation of new aspects of heuristic traffic grooming algorithms for green and low delay optical networks consisting of network devices with diverse energy-traffic proportionalities. This paper demonstrates that a heuristic traffic grooming algorithm enhanced by comparison with direct bypass routing improves both energy and delay performances in a high energy-traffic proportionality regime. A hottest-first sorting policy in flow provisioning can further improve delay performance even in a low energy-traffic proportionality regime.

Green IP over WDM network design considering energy-load proportionality

Considering the traffic explosion in the ICT sector, energy awareness in new technology development becomes one of the most critical requirement. Technology evolution in the semiconductor sector is not sufficient for a sustainable network growth by internet applications. We review importance of the energy proportionality in network device, which is one of the key factors in designing energy-aware network. This paper considers cross-layer optimization for energy consumption in the IP over WDM network. A passive optical component based optical node design is a promising technology with the negligible energy consumption in optical switching, but an IP router node is still power greedy. With ILP simulation to optimize the cross-layer design of IP and optical layers, we report that a traffic grooming technique exhibits different cross-layer design rules depending on the energy-load proportionality of a network device.

Traffic off-balancing algorithm: Toward energy proportional datacenter network

A traffic off-balancing algorithm for a 2-tiered datacenter network design is proposed and its energy savings gain is investigated with respect to energy proportionality and control granularity of devices.

Add-drop Benes network for scalable optical packet networks

For the scalable optical packet transport networks, an innovative design of add-drop Benes network (ADBN) is presented where the cost and energy consumption can be considerably reduced by element savings in the architecture. In a WDM optical packet transport switching node, the ADBNs are interconnected to achieve buffer sharing among multiple ADBNs. A corresponding switch configuration algorithm and architecture rules for the single ADBN and shared ADBN are proposed to mitigate the limited connection capability of the proposed ADBN designs. Switch scalability is verified in consideration of a crosstalk noise performance and element counts.

Energy efficient network planning: Issues and prospects in wired/wireless network

Energy aware technologies in Information and communications technology (ICT) are becoming one of the most exigent research realms for the sustainable network industry since energy efficiency from silicon-technologies is not enough to overcome an energy consumption trend due to the increase of network traffic. This paper emphasizes a significance of greening ICT researches with rigorous forecasts for future energy consumption estimations in ICT industry. We review the concept of energy proportionality index (EPI) as a major factor of energy-aware network planning. This paper also introduces fundamental research issues for green wired/wireless networks.

Traffic off-balancing algorithm for energy efficient networks

Physical layer of high-end network system uses multiple interface arrays. Under the load-balancing perspective, light load can be distributed to multiple interfaces. However, it can cause energy inefficiency in terms of the number of poor utilization interfaces. To tackle this energy inefficiency, traffic off-balancing algorithm for traffic adaptive interface sleep/awake is investigated. As a reference model, 40G/100G Ethernet is investigated. We report that suggested algorithm can achieve energy efficiency while satisfying traffic transmission requirement.

Efficient Design and Scalable Control for Store-and-Forward Capable Optical Transport Networks

The increasingly bursty nature of delaytolerable content delivery services requires new design and control perspectives in state-of-the-art end-to-end bufferless transport networks. Store-and-forward capable networked buffers in end-to-end lightpath networks enable an optical delay-tolerant networking (DTN) service that efficiently improves the capacity of optical networks by time-domain traffic shaping. We model an architecture of the optical DTN and asymptotically analyze the blocking probability performance as a function of the buffer node ratio. We propose a novel buffer node location algorithm by suggesting a bottleneckness centrality measure and evaluate the algorithm by comparing it with reference facility location algorithms. Rigorous simulation studies over practical core network topologies, as a function of networking cost, reveal higher performance of the optical DTN model as well as the proposed network-planning algorithms.

Scalable software-defined optical networking with high-performance routing and wavelength assignment algorithms

The feasibility of software-defined optical networking (SDON) for a practical application critically depends on scalability of centralized control performance. The paper, highly scalable routing and wavelength assignment (RWA) algorithms are investigated on an OpenFlow-based SDON testbed for proof-of-concept demonstration. Efficient RWA algorithms are proposed to achieve high performance in achieving network capacity with reduced computation cost, which is a significant attribute in a scalable centralized-control SDON. The proposed heuristic RWA algorithms differ in the orders of request processes and in the procedures of routing table updates. Combined in a shortest-path-based routing algorithm, a hottest-request-first processing policy that considers demand intensity and end-to-end distance information offers both the highest throughput of networks and acceptable computation scalability. We further investigate trade-off relationship between network throughput and computation complexity in routing table update procedure by a simulation study.