Joint Computing and Network Resource Scheduling in a Lambda Grid Network

Data-intensive Grid applications require huge data transfers between grid computing nodes. These computing nodes, where computing jobs are executed, are usually geographically separated. A grid network that employs optical wavelength division multiplexing (WDM) technology and optical switches to interconnect computing resources with dynamically provisioned multi-gigabit rate bandwidth lightpath is called a Lambda Grid network. A computing task may be executed on any one of several computing nodes which possesses the necessary resources. In order to reflect the reality in job scheduling, allocation of network resources for data transfer should be taken into consideration. However, few scheduling methods consider the communication contention on Lambda Grids. In this paper, we investigate the joint scheduling problem while considering both optical network and computing resources in a Lambda Grid network. The objective of our work is to maximize the total number of jobs that can be scheduled in a Lambda Grid network. An adaptive routing algorithm is proposed and implemented for accomplishing the communication tasks for every job submitted in the network. Four heuristics (FIFO, ESTF, LJF, RS) are implemented for job scheduling of the computational tasks. Simulation results prove the feasibility and efficiency of the proposed solution.

Joint Computing and Network Resource Scheduling in a Lambda Grid Network

Data-intensive Grid applications require huge data transfers between grid computing nodes. These computing nodes, where computing jobs are executed, are usually geographically separated. A grid network that employs optical wavelength division multiplexing (WDM) technology and optical switches to interconnect computing resources with dynamically provisioned multi-gigabit rate bandwidth lightpath is called a Lambda Grid network. A computing task may be executed on any one of several computing nodes which possesses the necessary resources. In order to reflect the reality in job scheduling, allocation of network resources for data transfer should be taken into consideration. However, few scheduling methods consider the communication contention on Lambda Grids. In this paper, we investigate the joint scheduling problem while considering both optical network and computing resources in a Lambda Grid network. The objective of our work is to maximize the total number of jobs that can be scheduled in a Lambda Grid network. An adaptive routing algorithm is proposed and implemented for accomplishing the communication tasks for every job submitted in the network. Four heuristics (FIFO, ESTF, LJF, RS) are implemented for job scheduling of the computational tasks. Simulation results prove the feasibility and efficiency of the proposed solution.

Virtual Topology Reconfiguration of Wavelength-routed Optical WDM Networks

The bandwidth requirements of the Internet are increasing every day and there are newer and more bandwidth-thirsty applications emerging on the horizon. Wavelength division multiplexing (WDM) is the next step towards leveraging the capabilities of the optical fiber, especially for wide-area backbone networks. The ability to switch a signal at intermediate nodes in a WDM network based on their wavelengths is known as wavelength-routing. One of the greatest advantages of using wavelength-routing WDM is the ability to create a virtual topology different from the physical topology of the underlying network. This virtual topology can be reconfigured when necessary, to improve performance. We discuss the previous work done on virtual topology design and also discuss and propose different reconfiguration algorithms applicable under different scenarios.

Integrated Intermediate Waveband and Wavelength Switching for Optical WDM Mesh Networks

As wavelength-division multiplexing (WDM) evolves towards practical applications in optical transport networks, waveband switching (WBS) has been introduced to cut down the operational costs and to reduce the complexities and sizes of network components, e.g., optical cross-connects (OXCs). This paper considers the routing, wavelength assignment and waveband assignment (RWWBA) problem in a WDM network supporting mixed waveband and wavelength switching. First, the techniques supporting waveband switching are studied, where a node architecture enabling mixed waveband and wavelength switching is proposed. Second, to solve the RWWBA problem with reduced switching costs and improved network throughput, the cost savings and call blocking probabilities along intermediate waveband-routes are analyzed. Our analysis reveals some important insights about the cost savings and call blocking probability in relation to the fiber capacity, the candidate path, and the traffic load. Third, based on our analysis, an online integrated intermediate WBS algorithm (IIWBS) is proposed. IIWBS determines the waveband switching route for a call along its candidate path according to the node connectivity, the link utilization, and the path length information. In addition, the IIWBS algorithm is adaptive to real network applications under dynamic traffic requests. Finally, our simulation results show that IIWBS outperforms a previous intermediate WBS algorithm and RWA algorithms in terms of network throughput and cost efficiency.

Approximation Algorithms for Survivable Multicommodity Flow Problems with Applications to Network Design

Multicommodity flow (MF) problems have a wide variety of applications in areas such as VLSI circuit design, network design, etc., and are therefore very well studied. The fractional MF problems are polynomial time solvable while integer versions are NP-complete. However, exact algorithms to solve the fractional MF problems have high computational complexity. Therefore approximation algorithms to solve the fractional MF problems have been explored in the literature to reduce their computational complexity. Using these approximation algorithms and the randomized rounding technique, polynomial time approximation algorithms have been explored in the literature. In the design of high-speed networks, such as optical wavelength division multiplexing (WDM) networks, providing survivability carries great significance. Survivability is the ability of the network to recover from failures. It further increases the complexity of network design and presents network designers with more formidable challenges. In this work we formulate the survivable versions of the MF problems. We build approximation algorithms for the survivable multicommodity flow (SMF) problems based on the framework of the approximation algorithms for the MF problems presented in [1] and [2]. We discuss applications of the SMF problems to solve survivable routing in capacitated networks.

Survivable Traffic Grooming with Path Protection at the Connection Level in WDM Mesh Networks

Survivable traffic grooming (STG) is a promising approach to provide reliable and resource-efficient multigranularity connection services in wavelength division multiplexing (WDM) optical networks. In this paper, we study the STG problem in WDM mesh optical networks employing path protection at the connection level. Both dedicated protection and shared protection schemes are considered. Given the network resources, the objective of the STG problem is to maximize network throughput. To enable survivability under various kinds of single failures such as fiber cut and duct cut, we consider the general shared risk link group (SRLG) diverse routing constraints. We first resort to the integer linear programming (ILP) approach to obtain optimal solutions. To address its high computational complexity, we then propose three efficient heuristics, namely separated survivable grooming algorithm (SSGA), integrated survivable grooming algorithm (ISGA) and tabu search survivable grooming algorithm (TSGA). While SSGA and ISGA correspond to an overlay network model and a peer network model respectively, TSGA further improves the grooming results from SSGA and ISGA by incorporating the effective tabu search method. Numerical results show that the heuristics achieve comparable solutions to the ILP approach, which uses significantly longer running times than the heuristics.

Rerouting Schemes for Dynamic Traffic Grooming in Optical WDM Mesh Networks

Traffic grooming in optical WDM mesh networks is a two-layer routing problem to effectively pack low-rate connections onto high-rate lightpaths, which, in turn, are established on wavelength links. In this work, we employ the rerouting approach to improve the network throughput under the dynamic traffic model. We propose two rerouting schemes, rerouting at lightpath level (RRAL) and rerouting at connection level (RRAC). A qualitative comparison is made between RRAL and RRAC. We also propose the critical-wavelength-avoiding one-lightpath-limited (CWA-1L) and critical-lightpath-avoiding one-connection-limited (CLA-1C) rerouting heuristics, which are based on the two rerouting schemes respectively. Simulation results show that rerouting reduces the connection blocking probability significantly.

A Novel Fiber Delay Line Buffering Architecture for Optical Packet Switching

Due to the lack of optical random access memory, optical fiber delay line (FDL) is currently the only way to implement optical buffering. Feed-forward and feedback are two kinds of FDL structures in optical buffering. Both have advantages and disadvantages. In this paper, we propose a more effective hybrid FDL architecture that combines the merits of both schemes. The core of this switch is the arrayed waveguide grating (AWG) and the tunable wavelength converter (TWC). It requires smaller optical device sizes and fewer wavelengths and has less noise than feedback architecture. At the same time, it can facilitate preemptive priority routing which feed-forward architecture cannot support. Our numerical results show that the new switch architecture significantly reduces packet loss probability.

Cross-Layer Packet Size Optimization for Wireless Terrestrial, Underwater, and Underground Sensor Networks

In this paper, a cross-layer solution for packet size optimization in wireless sensor networks (WSN) is introduced such that the effects of multi-hop routing, the broadcast nature of the physical wireless channel, and the effects of error control techniques are captured. A key result of this paper is that contrary to the conventional wireless networks, in wireless sensor networks, longer packets reduce the collision probability. Consequently, an optimization solution is formalized by using three different objective functions, i.e., packet throughput, energy consumption, and resource utilization. Furthermore, the effects of end-to-end latency and reliability constraints are investigated that may be required by a particular application. As a result, a generic, cross-layer optimization framework is developed to determine the optimal packet size in WSN. This framework is further extended to determine the optimal packet size in underwater and underground sensor networks. From this framework, the optimal packet sizes under various network parameters are determined.

Diversidade genética em germoplasma tropical de cebola estimada via marcadores RAPD

A cebola é uma cultura de expressiva importância socioeconômica para o Brasil. Marcantes contribuições para o desenvolvimento da cultura têm sido feitas utilizando-se germoplasma de cebola adaptado às regiões tropicais e subtropicais. Nesse contexto, o presente trabalho teve como objetivo estudar a diversidade genética existente em uma coleção de germoplasma potencialmente útil ao desenvolvimento de cultivares para essas regiões. Para isso, a variabilidade genética de um grupo de 21 acessos foi analisada via marcadores RAPD. Esses acessos ('Red Creole', 'Roxa IPA-3', 'Valenciana 14', 'Beta Cristal', 'Diamante', 'Composto IPA-6', 'Aurora', 'Bojuda Rio Grande', 'Alfa Tropical', 'Pêra IPA-4', 'Primavera', 'Belém IPA-9', 'Crioula Alto Vale', 'Conquista', 'Pira-Ouro', 'Vale-Ouro IPA-11', 'Franciscana IPA-10', 'Serrana', 'CNPH 6400', 'Petroline' e 'Baia Periforme') têm sido empregados como germoplasma e/ou foram desenvolvidos pelos programas de melhoramento genético de cebola conduzidos no Brasil. Dos 520 iniciadores ('primers') utilizados na triagem inicial, somente 38 confirmaram polimorfismos entre os 21 acessos. Esses 38 'primers' produziram 624 amplicons, dos quais 522 (83,7%) foram monomórficos e 102 (16,3%) polimórficos. Com base nos padrões revelados, seis grupos foram formados de acordo com a similaridade média global entre os acessos (= 0,72). Somente um desses seis grupos englobou mais de um acesso. O grupo principal (formado por 16 acessos) incluiu, predominantemente, as cultivares que apresentam no seu pedigree a contribuição de 'Baia Periforme' ('Diamante', 'Composto IPA-6', 'Aurora', 'Bojuda Rio Grande', 'Conquista', 'Pira-Ouro', 'Serrana', 'Vale-Ouro IPA-11', 'Baia Periforme', 'Primavera', 'Franciscana IPA-10', 'Belém IPA-9', 'Crioula Alto Vale', 'Petroline', 'Pêra IPA-4' e 'Alfa Tropical'). As cultivares 'Red Creole', 'Roxa IPA-3', 'Beta Cristal', 'CNPH 6400' e 'Valenciana 14' formaram agrupamentos isolados e distintos do grupo 'Baia Periforme', revelando, dessa forma, divergência genética entre essas cinco populações e o grupo principal. Verificou-se que os materiais estudados possuem base genética relativamente estreita, apresentando, em sua grande maioria origem na população 'Baia Periforme'. Existem, no entanto, alguns materiais divergentes, cuja diversidade pode ser explorada em programas de melhoramento.

Genotipagem de polimorfismos associados com sistemas de macho-esterilidade em acessos de cebola adaptados ao Brasil

A produção em escala comercial de sementes híbridas de cebola (Allium cepa) tem sido conduzida com o emprego de dois sistemas de macho-esterilidade do tipo genética-citoplasmática (CMS-S e CMS-T) em associação ao citoplasma normal (macho-fértil). No entanto, a análise molecular desses diferentes tipos citoplasmáticos ainda não está disponível para um grande número de acessos de cebola adaptados para cultivo em regiões tropicais. Além de adaptação às condições edafoclimáticas do Brasil, muitos desses acessos apresentam tolerância a doenças, sendo de potencial valor como genitores de híbridos. O presente trabalho visou identificar os tipos citoplasmáticos de acessos de cebola de diferentes grupos morfoagronômicos de interesse para o melhoramento genético no Brasil, usando a reação da polimerase em cadeia (PCR) com 'primers' específicos para regiões polimórficas do genoma mitocondrial de cebola. Foi observada, nos 66 acessos amostrados, a presença dos três principais tipos de citoplasma descritos para cebola (S, N e T). Foi constatada maior frequência do citoplasma S (56%) seguido do citoplasma T (25,8%). Em 18,2% das amostras, foi encontrado exclusivamente o citoplasma N. Essa caracterização pode ser útil para guiar a escolha de materiais genéticos dentro dos programas de melhoramento com objetivo de desenvolver cultivares híbridas adaptadas às condições tropicais.

QuMinS: fast and scalable querying, mining and summarizing multi-modal databases

Given a large image set, in which very few images have labels, how to guess labels for the remaining majority? How to spot images that need brand new labels different from the predefined ones? How to summarize these data to route the user’s attention to what really matters? Here we answer all these questions. Specifically, we propose QuMinS, a fast, scalable solution to two problems: (i) Low-labor labeling (LLL) – given an image set, very few images have labels, find the most appropriate labels for the rest; and (ii) Mining and attention routing – in the same setting, find clusters, the top-'N IND.O' outlier images, and the 'N IND.R' images that best represent the data. Experiments on satellite images spanning up to 2.25 GB show that, contrasting to the state-of-the-art labeling techniques, QuMinS scales linearly on the data size, being up to 40 times faster than top competitors (GCap), still achieving better or equal accuracy, it spots images that potentially require unpredicted labels, and it works even with tiny initial label sets, i.e., nearly five examples. We also report a case study of our method’s practical usage to show that QuMinS is a viable tool for automatic coffee crop detection from remote sensing images.

Design and control techniques of optical networks

The world of communication has changed quickly in the last decade resulting in the the rapid increase in the pace of peoples’ lives. This is due to the explosion of mobile communication and the internet which has now reached all levels of society. With such pressure for access to communication there is increased demand for bandwidth. Photonic technology is the right solution for high speed networks that have to supply wide bandwidth to new communication service providers. In particular this Ph.D. dissertation deals with DWDM optical packet-switched networks. The issue introduces a huge quantity of problems from physical layer up to transport layer. Here this subject is tackled from the network level perspective. The long term solution represented by optical packet switching has been fully explored in this years together with the Network Research Group at the department of Electronics, Computer Science and System of the University of Bologna. Some national as well as international projects supported this research like the Network of Excellence (NoE) e-Photon/ONe, funded by the European Commission in the Sixth Framework Programme and INTREPIDO project (End-to-end Traffic Engineering and Protection for IP over DWDM Optical Networks) funded by the Italian Ministry of Education, University and Scientific Research. Optical packet switching for DWDM networks is studied at single node level as well as at network level. In particular the techniques discussed are thought to be implemented for a long-haul transport network that connects local and metropolitan networks around the world. The main issues faced are contention resolution in a asynchronous variable packet length environment, adaptive routing, wavelength conversion and node architecture. Characteristics that a network must assure as quality of service and resilience are also explored at both node and network level. Results are mainly evaluated via simulation and through analysis.

Cross-layer optimizations in multi-hop ad hoc networks

Unlike traditional wireless networks, characterized by the presence of last-mile, static and reliable infrastructures, Mobile ad Hoc Networks (MANETs) are dynamically formed by collections of mobile and static terminals that exchange data by enabling each other's communication. Supporting multi-hop communication in a MANET is a challenging research area because it requires cooperation between different protocol layers (MAC, routing, transport). In particular, MAC and routing protocols could be considered mutually cooperative protocol layers. When a route is established, the exposed and hidden terminal problems at MAC layer may decrease the end-to-end performance proportionally with the length of each route. Conversely, the contention at MAC layer may cause a routing protocol to respond by initiating new routes queries and routing table updates. Multi-hop communication may also benefit the presence of pseudo-centralized virtual infrastructures obtained by grouping nodes into clusters. Clustering structures may facilitate the spatial reuse of resources by increasing the system capacity: at the same time, the clustering hierarchy may be used to coordinate transmissions events inside the network and to support intra-cluster routing schemes. Again, MAC and clustering protocols could be considered mutually cooperative protocol layers: the clustering scheme could support MAC layer coordination among nodes, by shifting the distributed MAC paradigm towards a pseudo-centralized MAC paradigm. On the other hand, the system benefits of the clustering scheme could be emphasized by the pseudo-centralized MAC layer with the support for differentiated access priorities and controlled contention. In this thesis, we propose cross-layer solutions involving joint design of MAC, clustering and routing protocols in MANETs. As main contribution, we study and analyze the integration of MAC and clustering schemes to support multi-hop communication in large-scale ad hoc networks. A novel clustering protocol, named Availability Clustering (AC), is defined under general nodes' heterogeneity assumptions in terms of connectivity, available energy and relative mobility. On this basis, we design and analyze a distributed and adaptive MAC protocol, named Differentiated Distributed Coordination Function (DDCF), whose focus is to implement adaptive access differentiation based on the node roles, which have been assigned by the upper-layer's clustering scheme. We extensively simulate the proposed clustering scheme by showing its effectiveness in dominating the network dynamics, under some stressing mobility models and different mobility rates. Based on these results, we propose a possible application of the cross-layer MAC+Clustering scheme to support the fast propagation of alert messages in a vehicular environment. At the same time, we investigate the integration of MAC and routing protocols in large scale multi-hop ad-hoc networks. A novel multipath routing scheme is proposed, by extending the AOMDV protocol with a novel load-balancing approach to concurrently distribute the traffic among the multiple paths. We also study the composition effect of a IEEE 802.11-based enhanced MAC forwarding mechanism called Fast Forward (FF), used to reduce the effects of self-contention among frames at the MAC layer. The protocol framework is modelled and extensively simulated for a large set of metrics and scenarios. For both the schemes, the simulation results reveal the benefits of the cross-layer MAC+routing and MAC+clustering approaches over single-layer solutions.

Interconnection systems for highly integrated computation devices

The sustained demand for faster,more powerful chips has beenmet by the availability of chip manufacturing processes allowing for the integration of increasing numbers of computation units onto a single die. The resulting outcome, especially in the embedded domain, has often been called SYSTEM-ON-CHIP (SOC) or MULTI-PROCESSOR SYSTEM-ON-CHIP (MPSOC). MPSoC design brings to the foreground a large number of challenges, one of the most prominent of which is the design of the chip interconnection. With a number of on-chip blocks presently ranging in the tens, and quickly approaching the hundreds, the novel issue of how to best provide on-chip communication resources is clearly felt. NETWORKS-ON-CHIPS (NOCS) are the most comprehensive and scalable answer to this design concern. By bringing large-scale networking concepts to the on-chip domain, they guarantee a structured answer to present and future communication requirements. The point-to-point connection and packet switching paradigms they involve are also of great help in minimizing wiring overhead and physical routing issues. However, as with any technology of recent inception, NoC design is still an evolving discipline. Several main areas of interest require deep investigation for NoCs to become viable solutions: • The design of the NoC architecture needs to strike the best tradeoff among performance, features and the tight area and power constraints of the on-chip domain. • Simulation and verification infrastructure must be put in place to explore, validate and optimize the NoC performance. • NoCs offer a huge design space, thanks to their extreme customizability in terms of topology and architectural parameters. Design tools are needed to prune this space and pick the best solutions. • Even more so given their global, distributed nature, it is essential to evaluate the physical implementation of NoCs to evaluate their suitability for next-generation designs and their area and power costs. This dissertation focuses on all of the above points, by describing a NoC architectural implementation called ×pipes; a NoC simulation environment within a cycle-accurate MPSoC emulator called MPARM; a NoC design flow consisting of a front-end tool for optimal NoC instantiation, called SunFloor, and a set of back-end facilities for the study of NoC physical implementations. This dissertation proves the viability of NoCs for current and upcoming designs, by outlining their advantages (alongwith a fewtradeoffs) and by providing a full NoC implementation framework. It also presents some examples of additional extensions of NoCs, allowing e.g. for increased fault tolerance, and outlines where NoCsmay find further application scenarios, such as in stacked chips.