O presente trabalho apresenta a implementação de um core de rede Ethernet 10/100Mbps com interface para o barramento Avalon para utilização em conjunto com o processador Nios II da Altera. A tecnologia Ethernet foi implementada em computação reconfigurável e utilizou-se como base um módulo disponível na Internet denominado OpenCores MAC 10/100. O projeto foi desenvolvido para ser aplicado em sistemas embarcados, mais especificamente para o uso em um robô móvel em desenvolvimento no Laboratório de Computação Reconfigurável do ICMC/USP. O core foi incorporado à biblioteca da ferramenta SoPC Builder da Altera, visando uma fácil integração do mesmo em outros projetos. Foram utilizadas as ferramentas Quartus II e ModelSim para o desenvolvimento e testes do sistema, além de dois kits Nios versão Stratix para a validação do projeto, sendo as placas interligadas ponto-a-ponto sem a utilizaçao de transceivers analógicos. / This work presents the implementation of a network Ethernet 10/100Mbps core with interfaces to Avalon bus for using with the Nios II processor from Altera. The Ethernet technology was implemented in reconfigurable computing and was based in the OpenCores MAC 10/100 available on Internet. The project was developed for embedded systems applications, more specifically for a mobile robot in development at Reconfigurable Computing Laboratory from ICMC/USP. The core was incorporated to SoPC Builder tool’s library from Altera, aiming to facilitate the integration with others projects. To development and system tests were used Quartus II and ModelSim, and two Nios Development kit Statix Edition for project validation. The boards were linked peer-to-peer, without use analog transceivers.

<p>Server-based architectures have recently generated more interests and are currently considered for usage for communication in networks. In parallel, switched Ethernet technology has been widely adopted and used in lots of networked systems. Current requirements of networks for supporting real-time guarantees while being flexible at the same time have made the network designers to consider addition of some features to common switches. The FTT-Enabled Ethernet switch is a switch that has been developed to support the FTT (Flexible Time Triggered) paradigm. Recently, servers have been added in these types of switches in order to efficiently manage their allocated bandwidth to different types of messages.</p><p>A hierarchical network of Ethernet switches might be designed in different ways according to the overall goals and properties of the network. In this thesis, after a study on different design solutions, an architecture has been proposed based on FTT-enabled switches, motivated by their support of real-time constraints and server-based communication features. After having created the architecture, a protocol for bandwidth reservation for this hierarchically composed Ethernet switch architecture is developed. Behavior of the designed protocol is described in detail and it has been modeled using Uppaal. Moreover, the temporal behavior (timing) of the network is presented.</p>

Server-based architectures have recently generated more interests and are currently considered for usage for communication in networks. In parallel, switched Ethernet technology has been widely adopted and used in lots of networked systems. Current requirements of networks for supporting real-time guarantees while being flexible at the same time have made the network designers to consider addition of some features to common switches. The FTT-Enabled Ethernet switch is a switch that has been developed to support the FTT (Flexible Time Triggered) paradigm. Recently, servers have been added in these types of switches in order to efficiently manage their allocated bandwidth to different types of messages. A hierarchical network of Ethernet switches might be designed in different ways according to the overall goals and properties of the network. In this thesis, after a study on different design solutions, an architecture has been proposed based on FTT-enabled switches, motivated by their support of real-time constraints and server-based communication features. After having created the architecture, a protocol for bandwidth reservation for this hierarchically composed Ethernet switch architecture is developed. Behavior of the designed protocol is described in detail and it has been modeled using Uppaal. Moreover, the temporal behavior (timing) of the network is presented.

This master’s thesis is based on the idea to extend a point-to-point 10 Gbps Ethernet cryptoto a multi-point 10 Gbps Ethernet crypto. The multi-point extension is targeted for a fictiveproduct called Krypet. Extending Krypet to multi-point can be seen as creating a distributed Ethernet switch wherethe ports are Krypet units and the information passed between the ports is encrypted. Problems that occur when this extension is designed are identified and analysed. Some problemexamples are address lookup, session changes and packet switching. Different solutions to the identified problems are presented and these are compared anddiscussed. Requirements on the whole multi-point extension are determined and an architecturebased on the solutions is designed. The architecture is implemented and the resultof the implementation is discussed. The implementation successfully extends the system tomulti-point support and satisfies most of the determined requirements.

Vorgestellt wird die Architektur des Data Centers der TU Chemnitz, die auf den neuen Technologie Data Center Bridging (DCB) und Fibre Channel over Ethernet (FCoE) basiert. Es werden die entsprechenden Standards dargestellt und ein Überblick zur gegenwärtig verfügbaren Technik gegeben. Das Rechenzentrum der TU Chemnitz setzt diese Technologien bereits erfolgreich im Kontext von VMware-Virtualisierung und bei Betrieb I/O-intensiver Systeme ein.

Lietuvoje kaip ir visame pasaulyje sparčiai vyksta elektronikos raida, įskaitant ir patalpų mikroklimato stebėjimą. Šiuolaikinės technologijos leidžia ne tik skaitmeniniu būdu gauti duomenis iš jutiklių, bet ir nuosekliai stebėti, juos išsaugoti norimu intervalu bei panaudoti mikroklimato įrenginių valdymui. Šiam darbui atlikti naudojami patalpų mikroklimato duomenų stebėjimo įtaisai. Elektronikos bakalauro baigiamojo darbo tema - "Patalpos mikroklimato duomenų surinkimo įrenginys". Pagrindinė darbo užduotis - suprojektuoti ir pagaminti įrenginio modelį. Atlikus panašių projektų ir technologijų apžvalgą, buvo parinkta toks įrenginio veikimo principas: informacijai apie oro sąlygas surinkti naudojami elektroniniai, skaitmeniniai jutikliai; parametrų registracijai pasirinkta atminties kortelė; realaus laiko stebėjimui internetu, visi duomenis matomi Web puslapyje, ethernet'o pagalba. / The entire countries world wide including Lithuania is undergoing rapid evolution of electronic development, including indoor microclimate monitoring. Today’s modern technology allows not only receive digital data from sensors, but also to consistently monitor it, maintain the desired data storing period and use it for climate control management. Indoor climate data monitoring devices were used for this work. Electronics Bachelor thesis topic is – “Indoor microclimate data logger”. The main task of the work is – to design and manufacture this model device. Following similar projects and technology reviews this device operating principle was chosen: electronic, digital sensors are used to collect information about weather condition; memory card is used to register parameters; real-time monitoring is done using internet, all data appears in Web page by Ethernet support.

Με την ανάπτυξη της τεχνολογίας οπτικών ινών κατά την τελευταία δεκαετία, η υποδομή των τηλεπικοινωνιών έχει παρουσιάσει μία τρομερή ανάπτυξη στη χωρητικότητα εύρους ζώνης των δικτύων κορμού. Η ανάπτυξη αυτή ξεκίνησε με τα δίκτυα ευρείας περιοχής (WANs) που παρέχουν συνδεσιμότητα μεταξύ πόλεων μέσω των δικτύων μητροπολιτικής περιοχής (MANs), τα οποία συνδέουν τα τηλεπικοινωνιακά κέντρα των παρόχων υπηρεσιών. Ωστόσο, τα δίκτυα πρόσβασης, που εξυπηρετούν πολυάριθμους οικιακούς χρήστες, μικρές επιχειρήσεις και οργανισμούς, δεν έχουν αναπτυχθεί ανάλογα. Παρόλο που η πρόσφατη ανάπτυξη των πρωτοποριακών τεχνολογιών xDSL και CaTV έχει αναβαθμίσει σημαντικά το τμήμα αυτό, οι τεχνολογίες αυτές είναι ακόμη ανεπαρκείς για να καλύψουν την ολοένα και αυξανόμενη ζήτηση εύρους ζώνης των χρηστών. Για την αντιμετώπιση αυτού του προβλήματος, οι τεχνολογίες της οπτικής ίνας μέχρι το σπίτι/το πεζοδρόμιο/το κτίριο (Fiber To The Home/Curb/Building – FTTH/FTTC/FTTB) έχουν θεωρηθεί από καιρό ως μια προτιμώμενη λύση και τα παθητικά οπτικά δίκτυα (PONs) θεωρούνται ευρέως ως μία πολλά υποσχόμενη τεχνολογία για την εφαρμογή διάφορων λύσεων FTTx. Η παρούσα διπλωματική εργασία αρχικά εισάγει το πρόβλημα σχετικά με την ανάγκη εφαρμογής οπτικών δικτύων πρόσβασης. Στη συνέχεια, εξετάζει πιθανές λύσεις στο πρόβλημα αυτό και αναδεικνύει τα παθητικά οπτικά δίκτυα (PONs) ως την καλύτερη επιλογή. Επιπλέον, γίνεται αναφορά στις διάφορες τεχνολογίες πολλαπλής πρόσβασης στα PONs (TDMA, OCDMA και WDMA). Η εφαρμογή της TDMA τεχνολογίας στα PONs αποδεικνύεται η πιο οικονομικά αποδοτική. Για τη χρήση της TDMA τεχνολογίας στα PONs αναπτύχθηκαν τα πρότυπα ATM PON (BPON), Ethernet PON (EPON) και Gigabit PON (GPON). Μετά από σχετική σύγκριση των προτύπων αυτών, το EPON θεωρείται η καλύτερη επιλογή για την εφαρμογή της TDMA τεχνολογίας στα PONs. Αυτή η διπλωματική εργασία στη συνέχεια εισάγει τις θεμελιώδεις έννοιες σχετικά με τα EPONs, εξετάζει τα κύρια ζητήματα που σχετίζονται με τη διάθεση εύρους ζώνης του ανοδικού καναλιού (από τον τελικό χρήστη προς το τηλεπικοινωνιακό κέντρο) στα συστήματα EPONs και παρουσιάζει μία επισκόπηση των πιο πρόσφατων αλγορίθμων δυναμικής διάθεσης εύρους ζώνης (Dynamic Bandwidth Allocation – DBA) για ανοδικές μεταδόσεις στα EPONs. Τέλος, γίνεται περιγραφή και υλοποίηση ενός εκ των πρώτων DBA αλγορίθμων που προτάθηκαν για τη διάθεση του ανοδικού εύρους ζώνης στα EPONs, του αλγορίθμου IPACT (Interleave Polling with Adaptive Cycle Time – IPACT). Κατά την περιγραφή, η limited service του αλγορίθμου IPACT προκύπτει ως η καλύτερη, με βάση διάφορα κριτήρια, μεταξύ των διάφορων υπηρεσιών αυτού του DBA αλγορίθμου και αποτελεί σημείο αναφοράς για μεταγενέστερες δημοσιευμένες εργασίες που υπάρχουν στη βιβλιογραφία. Η υλοποίηση περιλαμβάνει προσομοίωση της limited service του αλγορίθμου IPACT με χρήση της γλώσσας προσομοίωσης SIMSCRIPT II.5 και εξαγωγή αναλυτικών αποτελεσμάτων από εξισώσεις που μοντελοποιούν τη συμπεριφορά του αλγορίθμου, με χρήση της γλώσσας προγραμματισμού C. Τα αποτελέσματα της ανάλυσης και της προσομοίωσης συγκρίνονται και προκύπτουν τα κατάλληλα συμπεράσματα. / With the development of fiber-optic technology over the last decade, the telecommunications infrastructure has shown a tremendous growth in the bandwidth capacity of backbone networks. This development began with the wide area networks (WANs) that provide connectivity between cities through the metropolitan area networks (MANs), which connect service providers’ central offices. However, access networks that serve numerous residential users, small businesses and organizations have not been developed accordingly. Although the recent development of innovative xDSL and CaTV technologies has significantly upgraded this segment, these technologies are still inadequate to meet the ever-increasing bandwidth demand of users. To address this problem, fiber to the home / curb / building (FTTH / FTTC / FTTB) technologies have long been regarded as a preferred solution, and passive optical networks (PONs) are widely regarded as a promising technology for implementing various FTTx solutions. This thesis first introduces the problem concerning the need for optical access networks. Then, examines possible solutions to this problem and highlights the passive optical networks (PONs) as the best option. Furthermore, a reference is made to the various multiple access technologies in PONs (TDMA, OCDMA and WDMA). The implementation of the TDMA technology in PONs proved the most cost-effective. The ATM PON (BPON), Ethernet PON (EPON) and Gigabit PON (GPON) standards have been developed for the use of the TDMA technology in PONs. Upon comparison of these standards, the EPON is considered the best option for the implementation of the TDMA technology in PONs. This thesis then introduces the fundamental concepts on EPONs, examines the major issues related to bandwidth allocation of the upstream channel (from the end user to the central office) in EPONs systems and presents a survey of the most recent dynamic bandwidth allocation (DBA) algorithms for upstream transmissions in EPONs. Finally, a description and an implementation are made of one of the first DBA algorithms proposed for upstream bandwidth allocation in EPONs, the IPACT (Interleave Polling with Adaptive Cycle Time - IPACT) algorithm. In the description, the limited service of the IPACT algorithm emerges as the best among the different services of this DBA algorithm according to various criteria and is a benchmark for subsequent published papers in the literature. The implementation includes the simulation of the limited service of the IPACT algorithm, using the SIMSCRIPT II.5 simulation language and extraction of analytical results from equations modeling the behavior of the algorithm, using the C programming language. The analysis and simulation results are compared and the appropriate conclusions arise.

Les travaux portent sur l’économie d’énergie dans le secteur des technologies de la communication et plus particulièrement dans les réseaux filaires. La technologie support de nos travaux est Ethernet qui historiquement utilisée dans les entreprises est actuellement déployée dans les réseaux d’accès et de coeur des opérateurs. Notre objectif est d’économiser de l’énergie par une mise en sommeil des liens Ethernet en s’appuyant sur des mécanismes standards aisément déployables. Pour ce faire nous modélisons et évaluons le mécanisme de mise en sommeil IEEE802.3az et confrontons notre modèle à l’expérimentation sur équipements. A partir du modèle de coût obtenu, nous proposons de mettre en place une ingénierie de trafic verte fonction de la charge qui dirige le trafic dans le réseau de façon à permettre aux liens de se mettre en phase de sommeil tout en préservant une qualité de service en évitant de créer des pertes de données dans le réseau par une concentration trop importante de trafic sur les liens. Nous distinguons plusieurs politiques d’ingénierie de complexité de mise en œuvre différentes, que nous évaluons dans plusieurs contextes. Les résultats obtenus permettent quasiment de doubler les gains obtenus par le standard IEEE802.3az. Nous étudions l’ingénierie dans un mode distribué à partir d’un protocole de routage, OSPF, et présentons une preuve de concept dans un mode centralisé avec une architecture SDN pour laquelle nous proposons l’utilisation du routage source par segment afin de réduire le trafic de contrôle. / This research has proposed a novel Ethernet-based global energy efficiency technology for the wired communications networks.In the development of the “green” technology, the IEEE 802.3az standard (i.e. referred to as LEGACY) was first applied to render dormant the non-utilized links to realize the local energy efficiency.To further lower the energy demand and thus achieve the global energy efficiency, this research put forward two additional algorithmic schemes(i.e. the EAGER and CARE green metrics)whereby the data are aggregated such that as many links as possible remain unutilized. The three energy-efficient metrics (LEGACY, EAGER and CARE) were individually applied to the distributed and centralized networks and subsequently evaluated.The experimental results showed that, by comparison, the CARE green metric outperforms the other two metrics (LEGACY and EAGER) with regard to the global energy saving in both the distributed and centralized wired networks.

Made available in DSpace on 2014-12-17T14:56:11Z (GMT). No. of bitstreams: 1 RicardoAMV.pdf: 949844 bytes, checksum: ba5839bf5dddd12b97b03d717da638fd (MD5) Previous issue date: 2006-11-30 / The Ethernet technology dominates the market of computer local networks. However, it was not been established as technology for industrial automation set, where the requirements demand determinism and real-time performance. Many solutions have been proposed to solve the problem of non-determinism, which are based mainly on TDMA (Time Division Multiple Access), Token Passing and Master-Slave. This work of research carries through measured of performance that allows to compare the behavior of the Ethernet nets when submitted with the transmissions of data on protocols UDP and RAW Ethernet, as well as, on three different types of Ethernet technologies. The objective is to identify to the alternative amongst the protocols and analyzed Ethernet technologies that offer to a more satisfactory support the nets of the industrial automation and distributed real-time application / A tecnologia Ethernet domina o mercado de redes locais de computadores. Entretanto, n?o se estabeleceu como tecnologia para a automa??o industrial, onde os requisitos exigem determinismo e desempenho de tempo real. Muitas solu??es foram propostas para resolver o problema do n?o determinismo, as quais s?o baseadas principalmente em TDMA (Time Division Multiple Access), Token Pass e Master-Slave. Este trabalho de pesquisa realiza medidas de desempenho que permite comparar o comportamento das redes Ethernet quando submetidas ?s transmiss?es de dados sobre os protocolos UDP e RAW Ethernet, bem como, sobre tr?s tipos diferentes de tecnologias Ethernet. O objetivo ? identificar a alternativa dentre os protocolos e tecnologias Ethernet analisadas que oferecem um suporte mais satisfat?rio ?s redes da automa??o industrial, e aplica??es de tempo real distribu?das

Esta dissertação descreve os métodos, medidas e análises feitas para otimizar a rede de comunicação de dados do Instituto de Física de São Carlos. As tecnologias e protocolos utilizados na rede são apresentados. Especial atenção é dada a análise do desempenho de VLANs utilizando inicialmente o protocolo LANE no núcleo ATM da rede. Neste caso a rede é composta de switches ATM e ATM-Ethernet. Medidas comparativas foram realizadas com a utilização da tecnologia Fast Ethernet no backbone, que possui uma capacidade de transmissão relativamente próxima da ATM de 155Mbps. Melhores resultados obtidos com a implementação de sub-redes maiores, reduzindo em um numero menor de sub-redes, são discutidos e apresentados. Análises estatísticas baseadas apenas no tempo de resposta da rede são apresentados para avaliar o desempenho das alterações efetuadas nas configurações da rede. Mesmo que o tráfego tenha aumentado muito durante esses anos e que vários serviços tenha sido agregados a esta rede, foi possível adequar o desempenho as novas necessidades beneficiando-se da evolução tecnológica que os equipamentos de rede de dados trouxeram ao IFSC. / This work describes the methods, measures and analyses performed aiming to optimize the data communication network from the Physics Institute of Sao Carlos as well as the technologies and protocols used in the network recently. Special care is given to the analysis of the VLANs performance using, initially, the LANE protocol over ATM which has its core based on pure ATM and ATM-Ethernet switches. Comparative measures had been carried out using a backbone working on a Fast Ethernet technology, which seems to have a very close transmission rate from the ATM 155Mbps. This work also discusses best results acquired with the implementation of larger networks reducing the number of subnetworks, statistical analyses based on time delay of the network in order to evaluate the performance of the changes made on its configuration. Even though the traffic from the Institute has been increased over these years and several services have been added to the network, it was completely possible to adapt the performance to the needs, using the benefits of the technological evolution which the network equipments had brought to the Institute.