Jeremy M. Plante

Manycasting: Energy-Efficient Multicasting in WDM Optical Unicast Networks

With the increasing number of high-bandwidth applications, energy consumption of networks has become an important issue that needs to be addressed. Manycasting is a communication paradigm that finds applications in such high-bandwidth environments. To support manycasting functionality in an optical network that is Split-Incapable (SI), i.e., the optical cross connects are incapable of switching an incoming optical signal to more than one output interface, manycasting must be implemented as an overlay to the optical layer. We propose two such overlay approaches: Manycasting with Drop at Member Node (MA-DAMN) and Manycasting with Drop at Any Node (MA-DAAN) which employ Steiner tree routing. We subject these approaches to a static traffic model, and present integer linear programs (ILPs) and heuristic approximations which aim to minimize the total number of wavelengths required to service the manycast request set in an effort to reduce network-wide energy consumption. Through extensive simulation, we show that MA-DAMN and MA-DAAN achieve 25−45% wavelength reduction as compared to a baseline overlay approach which employs single-hop lightpaths to each manycast destination individually.

Renewable Energy-Aware Manycast Overlays

Manycasting allows a single source to reach multiple destinations while providing flexibility in destination selection. Our goal in this paper is to improve the cost of the manycast drop at member node (MA-DMN) overlay algorithm in terms of energy consumption and associated greenhouse gas (GHG) emissions. To reduce the environmental impact, ideally, a large percentage of the network nodes along the transmission and the chosen destinations need to be green. We present a novel energy-conservative emission-aware variant of the MA-DMN algorithm. We then propose further modifications to increase the utilization of those destinations that are powered by renewable energy sources: manycast drop at greenest nodes (MA-DGN). The potential for emission reduction by those algorithms is two-fold: The data are transported in the most efficient way and processed at the greenest available data centers. We compare the approaches by simulating realistic quantities of dynamic traffic. We assume heterogeneously distributed and time-dependent availability of renewable energy sources to power nodes throughout the network. We find that the energy-source-aware algorithms lower both energy-consumption and GHG emissions at stable network performance levels, in some cases even lowers blocking rate.

Crosstalk-aware anycast routing and wavelength assignment in optical WDM networks

In this paper we discuss the performance of physical-layer impairment-aware anycast communication over transparent optical networks. High-bandwidth applications, such as grid computing over optical networks will benefit from using anycast requests. From the simulation results we observe that the proposed anycast routing algorithms can significantly decrease the blocking probability of requests due to impairments, such as crosstalk and ASE noise. 1

On extending ESnet's OSCARS with a multi-domain anycast service

Current scientific data applications require advanced network provisioning systems to support the transport of large volumes of data. Due to the availability of diverse computing and Grid clusters, these applications can benefit from anycasting capabilities. In contrast to unicasting, anycast routing allows the selection of a node from a group of candidate destinations. This new means of communication allows for greater routing flexibility and better network resource consumption. However, current provisioning systems do not provide fully compliant anycast implementations. In this paper, we extend ESnets OSCARS virtual circuit provisioning system with anycast routing capabilities to support destination-agnostic applications on single- and multi-domain network scenarios. The proposed implementation significantly improves provisioning success over the native unicast implementation in compliance with the existing OSCARS framework.

Energy source-aware manycast overlay in WDM networks

Manycasting is an emerging communication paradigm which allows a single source to reach multiple destinations while providing flexibility in the selection of which destinations to connect with. Traditional wavelength division multiplexed (WDM) networks do not support the all-optical splitting of signals to multiple output ports as required by point-to-multipoint communication schemes. Previous work has proposed an overlay approach known as Manycasting with Drop at Member Node (MA-DMN) to provide manycast support as a logical overlay to basic point-to-point lightpath connections. This approach has been studied extensively and compared to alternative overlay models, and has emerged the obvious candidate for supporting manycast overlays. Throughout its evaluation though, MA-DMN has never been scrutinized in terms of its costs for energy consumption and associated greenhouse gas (GHG) emissions. In this work, we subject MA-DMN to these evaluations, while also proposing a new more energy-conservative emission-aware variant known as MA-DMN using Least Impact Trees (MA-DMN-LIT). We compare these two approaches by simulating realistic quantities of dynamic traffic, and uniformly distributing renewable energy sources to power nodes throughout the network. We find that MA-DMN-LIT reduces energy consumption over MA-DMN by 6-10% across the network, while also reducing CO2 emissions by as much as 27%. We further conclude that MA-DMN-LIT also provides lower connection blocking by not over-subscribing shorter paths in the network as its emission-blind counterpart does.

Dynamic Multicasting in WDM Optical Unicast Networks for Bandwidth-Intensive Applications

Traditionally in wavelength division multiplexed (WDM) networks, multicasting is accomplished by splitting the signals all-optically, thereby establishing a tree of lightpaths (light-tree) from the source to every destination. To provision for this functionality in a Multicast-Incapable environment, in which the switches are not capable of directing an incoming signal to more than one output interface, one must implement a logical multicast overlay to the underlying optical layer. A naive method of accomplishing this is by creating a set of unicast lightpaths from the source to each destination of the multicast request. However, for large multicast groups, this leads to a poor utilization of the network resources. To alleviate this problem, we present two multicast Steiner tree overlay alternatives: Multicast with Drop At Member Node (DAMN), in which a lightpath may only terminate at member nodes of the multicast request, and Multicast with Drop At Any Node (DAAN), in which lightpaths may terminate at any node in the physical topology. We consider a dynamic traffic model, and propose efficient heuristics to solve the DAMN and DAAN problems with a goal of minimizing the total number of wavelengths required to satisfy the request. Moreover, we present a simple heuristic to approximate the baseline unicast approach (naive method). Our results demonstrate that at various loads, both the DAMN and DAAN reduce wavelength consumption by 42 - 60% over the naive unicast approach in realistic networks.

Critical resource multicast protection in data center networks

Resources in a network are imperfect, and equipment failure can have detrimental effects on data and transmission success rates. Attempts to improve the survivability of network communications when these failures occur focus primarily on protection against the common occurrence of link failures, while nodal failure has been largely overlooked. In data-critical infrastructures, such as cloud computing or data center networks, wherein the critical points of interest are at the nodes, a natural disaster or directed attack could have catastrophic consequences for the localized data. To overcome single points of failure, replicated multicast transmissions can be used to distribute copies of critical data to various geographically dispersed locations on the grid. We therefore explore strategies to protect against single critical node failures during multicast transmissions. We propose three novel multicast survivable heuristics and quantitatively analyze and then compare them to traditional multicast provisioning schemes through extensive simulation1.

Slotted advance reservation for multicast-incapable optical wavelength division multiplexing networks

In this paper we investigate techniques for provisioning advance reservation (AR) multicast requests in multicast-incapable (MI) networks, which lack the ability to split an incoming signal to multiple output ports, without performing an O-E-O conversion. AR traffic consists of connection requests that arrive and reserve network resources at some time before they need them to ensure better qualityof service than on-demand requests would receive. The traditional approach of providing multicast support in MI networks is to use an overlay approach in which a set of lightpaths is established from the source to each multicast destination member independently. This approach is wasteful of wavelength resources, particularly as the multicast destination set grows. We propose two alternative overlay approaches that take advantage of multiple-hop overlay-tree structures to limit the consumption of wavelengths in the network. We investigate static traffic scenarios on various network topologies and develop integer linear programs (ILPs) to optimally solve all three of the overlay-tree problems presented in this work with the goal of minimizing the total number of wavelengths required to service a multicast request set. We also present efficient heuristics that build and select overlay-trees that lower dynamic connection blocking and wavelength consumption. We compare the heuristics to the optimal ILPs on a small-scale network, and then further evaluate the heuristics on several large-scale topologies. In all scenarios, we are able to conclude that by sacrificing a minimization of O-E-O conversions, our more flexible overlay approaches, called drop at member node (MI-DMN) and dropatany node (MI-DAN), are superiorinterms of resource usage when compared with the traditional naive approach. Further dynamic traffic evaluations reveal that blocking may be lowered over the naive approach by more thantwoordersofmagnitudeatlowtomedium traffic loads.

Parallel circuit provisioning in ESnet's OSCARS

Large-scale science applications generate great volumes of data, which are frequently stored in remote data repositories or shared with cooperating laboratories across the network through the use of advance reservation connections. The groups that utilize these data transfers would benefit from having their applications simultaneously transmit data over multiple channels in parallel. Many of todays networks do not however provide the resource-aware scheduling to support this parallelization. As part of a framework for providing said applications with parallel resource-optimized provisioning of end-to-end requests, we propose and develop a scheduling enhancement to ESnets On-demand Secure Circuits and Advance Reservation System (OSCARS). This enhancement comes in the form of a front-end client, the behavior of which we quantitatively evaluate to compare the performance of parallel resource-provisioning to serial resource usage for both unicast and anycast scenarios.

Dynamic manycasting in optical split-incapable WDM networks for supporting high-bandwidth applications

With the advent of bandwidth intensive applications, the demand for manycast networking capabilities has become an essential component of wavelength division multiplexed (WDM) optical networks. Traditionally, the manycast functionality is accomplished by splitting a signal all-optically, thereby creating a light-tree, which originates from the source node and reaches a subset of the destination nodes. To support the manycasting functionality in an optical network that is split-incapable (SI), i.e., the optical crossconnects are incapable of switching an incoming optical signal to more than one output interface, one must implement a logical overlay to the underlying optical layer. A naiıve approach to accomplish this is by creating a set of unicast lightpaths that originate at the source node and terminate at a subset of the destination nodes of the manycast request. We refer to this as the manycasting via WDM unicast (MA-VWU) approach. However, for a large number of requests this approach leads to a poor utilization of network resources. To overcome this problem, we propose two overlay approaches: manycasting with drop at member node (MA-DAMN) and manycasting with drop at any node (MA-DAAN). In these solutions, we achieve manycasting by creating a set of lightpath routes (possibly multiple hops) in the overlay layer. We consider a dynamic traffic model and propose efficient heuristics to solve the MA-DAMN and MA-DAAN problems with a goal of minimizing the total number of wavelengths required to satisfy the requests. Our results demonstrate that both the overlay approaches reduce the wavelength consumption by approximately 33-45% over the MA-VWU approach for real-world large-scale networks1.