The increasing awareness of the impact of the IT sector on the environment, together with economic factors, have fueled many research efforts to reduce the energy expenditure of data centers. Recent work proposes to achieve additional energy savings by exploiting, in concert with customers, service workloads and to reduce data centers’ carbon footprints by adopting demand-response mechanisms between data centers and their energy providers. In this paper, we debate about the incentives that customers and data centers can have to adopt such measures and propose a new service type and pricing scheme that is economically attractive and technically realizable. Simulation results based on real measurements confirm that our scheme can achieve additional energy savings while preserving service performance and the interests of data centers and customers.
This work describes a novel approach for the reduction of energy consumption in data centres (DCs) that will yield benefits both in terms of running costs and its environmental impact. The method is based on the introduction of collaborative interactions and flexibility clauses in contracts between all the DC ecosystem entities. The included entities are all the actors along the energy
production–consumption chain, from the energy provider to the Information Technology customer. The collaborative approach also integrates the interaction between federated DCs. In this paper, we find a detailed description of the architecture that enables interaction between the DC ecosystem parties, which is designed to be progressively deployed, allowing traditional and ‘greened’ services to coexist, and without modification of the existing DC automation and framework systems.
This letter proposes a general packet delivery
mechanism in which routing decisions are taken on the basis
of the identifiers assigned to each node. Such identifiers depend
on the position of each node in the network, and their purpose
is to simplify as much as possible the communication process
between each pair of nodes. The proposal is based on building
trees as the Routing Protocol for Low-Power and Lossy Networks
(RPL), described in RFC 6550, does. Then, each node of the
tree receives a unique identifier following a hierarchical scheme.
Hence, every two nodes of the tree can communicate with each
other performing simple bitwise XOR operations at intermediate
nodes instead of building, storing and maintaining complex
Botero, J.; Rincon, D.; Agusti, A.; Hesselbach, X.; Raspall, F.; Remondo, D.; Barba, A.; Barone, P.; Giuliani, G. International Workshop on Energy-Efficient Data Centres p. 1-12 Presentation's date: 2013-05-21 Presentation of work at congresses
In recent years, the emergence of the cloud computing has increased the need of resources to support cloud-based services. Therefore, the role of the data centers has become essential. Following the growing of services, the power consumption has increased dramatically, while the need for energy savings and CO2 reduction has become a requirement for a sustainable world.
The All4Green project fosters collaboration between energy providers (EP), data centers (DC) and customers/end users (EU) in order to provide energy sav-ings and CO2 emissions reduction. In this architecture, the contract binding EPs and DCs includes flexibility terms in order to allow the collaboration in the form of discounts that can be transferred also to DC customers, if they are willing to collaborate.
This work proposes an analytical model to characterize the performance of a loss-free
transmission strategy for Optical Burst and Packet Switched Networks, which ensures that
a burst/packet will successfully reach its destination once it has been scheduled in the first
link of its path. In this paper this strategy is called Sent-But-Sure (SBS) because it avoids
losses in any intermediate node. The SBS strategy combines a routing and wavelength
assignment scheme with simple contention resolution mechanisms. As a result, new
burst/packet allocation attempts in an intermediate node only contend with bursts/packets
in transit coming from a single input link. Moreover, bursts/packets in transit always have
priority over bursts/packets whose transmission has not been scheduled yet. These two
main features of the SBS strategy allow us to develop an analytical model based on a twopriority
M/G/1 queueing model to characterize the network performance.
Remondo, D.; Nunes, M.; Sargento, S.; Cesana, M.; Filippini, I.; Triay, J.; Agusti, A.; de Andrade, M.; Gutierrez, L.; Sallent, S.; Cervelló-Pastor, C. Conference on Next Generation Internet Design and Engineering p. 1-8 DOI: 10.1109/NGI.2009.5175764 Presentation's date: 2009-07-03 Presentation of work at congresses
Loss-free schemes are defined to ensure successful packet/burst transmissions in optical packet/burst
switching networks. To this end, they rely on a collision-free routing and wavelength assignment (CF-RWA)
scheme combined with simple contention resolution mechanisms that guarantee the absence of losses in
intermediate links. Here, the CF-RWA problem is studied. In particular, by using graph theory, the problem of
finding CF-RWA schemes that minimise the number of wavelengths to serve a given traffic matrix is set. The
problem is simplified when it is formulated by using pre-defined sets of non-colliding paths. Within this
framework, the problem is shown to be equivalent to finding a given vertex-set colouring of the so-called
restriction digraph. Here, two heuristic algorithms are proposed to obtain such vertex-set colourings. One of
them provides a suitable CF-RWA without having to solve the minimisation problem. By way of example, the
proposed method is applied to the NSFNet and the EON network providing quasi-optimal results.
Contention resolution is a major issue in OBS networks. Several proposals that ensure burst transmissions without losses inside the network have been studied in the literature. Two of these proposals are based on combining a collision-free routing and wavelength assignment scheme with simple contention avoidance/resolution mechanisms. The static approach defines variable offsets and ensures contention avoidance by means of a suitable pre-assignment of offset windows to each communication. The dynamic approach guarantees the successful resolution of all contentions by using a single FDL at each intermediate node. Both proposals are based on giving priority to transmissions coming from the upstream. Hence, when an upstream node misbehaves or changes its transmission traffic pattern, it might delay new burst allocations on downstream nodes, leading to burst losses in the worst case. In this paper we deal with the fairness issue of these proposals. Thus, we introduce simple mechanisms that guarantee the transmission of the committed load for each communication, allowing, at the same time, the dynamic sharing of the spare bandwidth on each wavelength. The proposed mechanisms are analyzed by means of simulation.
Método general para el diseño de esquemas sin pérdidas en redes OBS.
Se propone una estrategia general de transmisión sin pérdidas para redes de conmutación óptica de ráfagas que combina un esquema de asignación de longitudes de onda y caminos con mecanismos simples de resolución de contiendas. La solución propuesta cumple la restricción de continuidad en longitud de onda, es aplicable en entornos de transmisión síncronos y asíncronos, y es independiente de la estadística de generación y de la longitud de las ráfagas. El patrón de transmisión que se define para cada longitud de onda asegura que toda contienda se puede resolver retardando de forma adecuada la transmisión de las ráfagas que compiten por el uso simultáneo de recursos en un enlace intermedio. Dicho retardo adicional (independiente de la longitud de los enlaces) se puede introducir en la programación de las ráfagas en el nodo origen (solución estática) o bien añadirse en un nodo intermedio cuando se produce contienda (solución dinámica).
The main objective of the Euro FGI network is to develop and maintain the most prominent European centre of excellence in Future Generation Internet (FGI) design and engineering, acting as a Collective Intelligence Think Tank, representing a major support for the European industry and leading towards a European leadership in this domain.
The FGI will offer multi-service/multimedia, mobility, service ubiquity and context awareness, convergence (services and fixed-mobile), Quality of Service, variable connectivity (always best connected), spontaneous networking and other capabilities as the norm.
Technology diversity is growing fast and mastering such a heterogeneous environment becomes essential for network designers. This new environment makes obsolete the design and engineering methods and tools currently available and forces the scientific community to develop new principles and methods to design/dimension/control/manage the new multi-technology architectures. These architectures will provide seamless end-to-end connectivity by hiding the technology diversity from service developers and users.
In addition, future high-speed wire-line and wireless access technologies provide instant high bandwidth connectivity making it difficult to forecast demand and thus to apply traditional traffic engineering methods.
For addressing this new environment, Euro-FGI will strengthen the integration of the scientific community activities to fulfil two main goals:
- Overcoming the challenge of technology diversity (vertical and horizontal integration) in the design of efficient and flexible FGI architectures
- Providing innovative traffic engineering approaches adapted to the new requirements and developing the appropriate quantitative methods for analysis, simulation and measurement
Due to the wealth of technologies and tools, achieving the above cited objectives requires the integration of a wide range of research capacities; a role that will be fulfilled by Euro-FGI.
Agusti, A.; Cervelló-Pastor, C.; Fiol, M. Sixth IEEE / CreateNetInternational Workshop on Optical Burst/Packet Switching (WOBS'06) held in conjunction with Third IEEE International Conference on Broadband Communications, Networks and Systems ( BROADNETS 2006) Presentation of work at congresses
Agusti, A.; Cervelló-Pastor, C. Fifth IEEE / CreateNet International Workshop on Optical Burst/Packet Switching (WOBS'05) held in conjunction with Second International Conference on Broadband Networks (BROADNETS 2005) p. 183-191 Presentation of work at congresses
The main objective of the Euro NGI network is to create and maintain the most prominent European centre of excellence in Next Generation Internet design and engineering, acting as a 'Collective Intelligence Think Tank', representing a major support for the European Information Society industry and leading towards a European leadership in this domain.
The recent technological advances will lead to exploitable innovative services once the integration of these technologies through innovative architectures is achieved. The Next Generation Internet will offer view multi-service/multimedia, mobility, convergence (services and fixed-mobile), Quality-of-Service and variable connectivity as the norm. On one hand, future high-speed wire-line and wireless access technologies provide instant high bandwidth connectivity, which makes difficult to forecast traffic and thus to apply existing traffic engineering methods. On the other hand, the technology diversity explodes and mastering such a heterogeneous environment becomes essential to the network designer. This requires investigation into new multi-technology architectures. It is understood today that new design, planning, dimensioning and management principles are needed.
In this context, the main topics addressed by the NoE are:
- Mastering the technology diversity (vertical and horizontal integration) for the design of efficient and flexible NGI architectures.
- Providing required innovative traffic engineering architectures adapted to the new requirements and developing the corresponding appropriate quantitative methods.
The network experts bring the required competences on the various technologies that will be integrated. They also bring a worldwide recognize expertise on the various topics that composes the traffic engineering and optimal dimensioning domain. Due to the wealth of technologies and tools, achieving the cited objectives requires the integration of European research capacities to reach a critical mass.