O uso de técnicas de cópia-zero e desvio do sistema operacional permitem a diminuição da latência de comunicação e o aumento da largura de banda. Menores latências e maiores larguras de banda contribuem para que o desempenho das aplicações paralelas seja mais alto, bem como torna-as mais escaláveis. Protocolos de comunicação que utilizam-se destas técnicas são conhecidos como protocolos de comunicação em nível de usuário. Baseado nas experiências de outros grupos de pesquisa na implementação de bibliotecas de comunicação e bibliotecas de programação paralelas sobre VIA e na experiência do GPPD na implementação da biblioteca DECK, este texto apresenta a implementação das primitivas DECK sobre o padrão VIA, o qual é classificado como sendo um protocolo de nível de usuário. O objetivo desta dissertação é implementar o DECK sobre VIA evitando qualquer cópia intermediária na comunicação de uma mensagem, atingindo assim cópia-zero. Dentre as bibliotecas de comunicação sobre VIA, DECK/VIA foi a única biblioteca que teve o compromisso ser totalmente livre de cópias intermediárias, embora houvesse que forçar um sincronismo na comunicação para manter este compromisso. Para a implementação do DECK/VIA, utilizou-se a implementação VI-GM de VIA para redes Myrinet. A biblioteca DECK/VIA demonstrou uma latência de 86.85 μs e uma largura de banda máxima de 205 Mbytes/s, 82% da banda nominal da rede Myrinet. Para validar a biblioteca foi executada a aplicação FT do pacote NPB. Apresenta-se comparações destes resultados frente aos resultados obtidos pela execução da mesma aplicação no DECK/GM, para redesMyrinet e DECK/TCP, para redes Ethernet. Constatou-se que mesmo com uma camada a mais de software e realizando todas as comunicações em três vias em virtude do handshake, DECK/VIA conseguiu valores de speedup bastante próximos de DECK/GM e de DECK/TCP para Gigabit Ethernet, superando os valores de DECK/TCP para Fast Ethernet. Conclui-se que o ideal na implementação de bibliotecas de programação paralela é encontrar uma solução balanceada entre a busca pelo desempenho e a manutenção da semântica original da biblioteca. O trabalho contribuiu com um survey de diversas soluções encontradas por outros grupos no desenvolvimento de bibliotecas de comunicação, que pode servir de guia para outros pesquisadores no desempenho da mesma tarefa. Também contribui com a introdução de um algoritmo para prevenção de deadlocks causados por comunicações síncronas. / Techniques like zero-copy and operating system bypass can decrease communication latency and increase bandwidth. Smaller latencies and greater bandwidths contribute for better performance in parallel applications and became them more scalables as well. Communication protocols using these techiniques are known as user-level communication protocols. Based on experiences from another research groups implementing communication libraries and parallel programming libraries over VIA and experience from GPPD implementing DECK, the text presents the implementation of DECK primitives over VIA standard, which is classified as an user-level protocol. The goal of this master’s thesis is implement DECK over VIA avoiding any intermediate copy between the data source and destination, reaching zero-copy. DECK/VIA is the unique library among all libriaries over VIA here studied totally free of intermediate copies, although a synchronous behavior was forced to keep this compromise. VI-GM, an implementation of VIA for Myrinet networks was used to implement DECK/VIA library. The implementation of DECK/VIA has shown a one-way latency of 86.85 μs and a maximum bandwidth of 205 Mbytes/s, 82% of nominal bandwidth of Myrinet network. To validate the library, the FT application from NPB was executed. Their results were compared with the results obtained with DECK/GM, for Myrinet networks and DECK/TCP, for Ethernet networks. Even with one additional software layer and doing all communication using a handshake, DECK/VIA reaches speedup values very closer of DECK/GMand DECK/TCP on Gigabit Ethernet and was better than DECK/TCP on Fast Ethernet. When implementing parallel programming libraries, we concluded the ideal solution is that meets the good balance between the quest for performance and the keeping of original library’s semantics. This work contibutes with a survey of communication libraries development, their problems and their solutions, which can guide others researchers performing the same task. Also it contributes with an algorithm to prevent deadlocks caused by synchonism.

O uso de técnicas de cópia-zero e desvio do sistema operacional permitem a diminuição da latência de comunicação e o aumento da largura de banda. Menores latências e maiores larguras de banda contribuem para que o desempenho das aplicações paralelas seja mais alto, bem como torna-as mais escaláveis. Protocolos de comunicação que utilizam-se destas técnicas são conhecidos como protocolos de comunicação em nível de usuário. Baseado nas experiências de outros grupos de pesquisa na implementação de bibliotecas de comunicação e bibliotecas de programação paralelas sobre VIA e na experiência do GPPD na implementação da biblioteca DECK, este texto apresenta a implementação das primitivas DECK sobre o padrão VIA, o qual é classificado como sendo um protocolo de nível de usuário. O objetivo desta dissertação é implementar o DECK sobre VIA evitando qualquer cópia intermediária na comunicação de uma mensagem, atingindo assim cópia-zero. Dentre as bibliotecas de comunicação sobre VIA, DECK/VIA foi a única biblioteca que teve o compromisso ser totalmente livre de cópias intermediárias, embora houvesse que forçar um sincronismo na comunicação para manter este compromisso. Para a implementação do DECK/VIA, utilizou-se a implementação VI-GM de VIA para redes Myrinet. A biblioteca DECK/VIA demonstrou uma latência de 86.85 μs e uma largura de banda máxima de 205 Mbytes/s, 82% da banda nominal da rede Myrinet. Para validar a biblioteca foi executada a aplicação FT do pacote NPB. Apresenta-se comparações destes resultados frente aos resultados obtidos pela execução da mesma aplicação no DECK/GM, para redesMyrinet e DECK/TCP, para redes Ethernet. Constatou-se que mesmo com uma camada a mais de software e realizando todas as comunicações em três vias em virtude do handshake, DECK/VIA conseguiu valores de speedup bastante próximos de DECK/GM e de DECK/TCP para Gigabit Ethernet, superando os valores de DECK/TCP para Fast Ethernet. Conclui-se que o ideal na implementação de bibliotecas de programação paralela é encontrar uma solução balanceada entre a busca pelo desempenho e a manutenção da semântica original da biblioteca. O trabalho contribuiu com um survey de diversas soluções encontradas por outros grupos no desenvolvimento de bibliotecas de comunicação, que pode servir de guia para outros pesquisadores no desempenho da mesma tarefa. Também contribui com a introdução de um algoritmo para prevenção de deadlocks causados por comunicações síncronas. / Techniques like zero-copy and operating system bypass can decrease communication latency and increase bandwidth. Smaller latencies and greater bandwidths contribute for better performance in parallel applications and became them more scalables as well. Communication protocols using these techiniques are known as user-level communication protocols. Based on experiences from another research groups implementing communication libraries and parallel programming libraries over VIA and experience from GPPD implementing DECK, the text presents the implementation of DECK primitives over VIA standard, which is classified as an user-level protocol. The goal of this master’s thesis is implement DECK over VIA avoiding any intermediate copy between the data source and destination, reaching zero-copy. DECK/VIA is the unique library among all libriaries over VIA here studied totally free of intermediate copies, although a synchronous behavior was forced to keep this compromise. VI-GM, an implementation of VIA for Myrinet networks was used to implement DECK/VIA library. The implementation of DECK/VIA has shown a one-way latency of 86.85 μs and a maximum bandwidth of 205 Mbytes/s, 82% of nominal bandwidth of Myrinet network. To validate the library, the FT application from NPB was executed. Their results were compared with the results obtained with DECK/GM, for Myrinet networks and DECK/TCP, for Ethernet networks. Even with one additional software layer and doing all communication using a handshake, DECK/VIA reaches speedup values very closer of DECK/GMand DECK/TCP on Gigabit Ethernet and was better than DECK/TCP on Fast Ethernet. When implementing parallel programming libraries, we concluded the ideal solution is that meets the good balance between the quest for performance and the keeping of original library’s semantics. This work contibutes with a survey of communication libraries development, their problems and their solutions, which can guide others researchers performing the same task. Also it contributes with an algorithm to prevent deadlocks caused by synchonism.

The thesis provides a basic concept of intelligent building and related fields where this technology is appropriate. The next section describes the systems used in this field, especially the so-called communication buses and protocols. For an overview of developments in this area there is also a list of major companies involved in this technology. As a practical example is then described by air temperature and air control facilities in the building A2 or A5 in a building located within the Faculty of Mechanical Engineering, Brno University of Technology. For both cases are also proposed possible treatment.

This paper describes systems and user interfaces for monitoring and supervisory control of technological equipment. It describes basic approaches used to solve this problem, commonly used communication protocols and existing user interfaces. Based on those findings are defined requirements for design of a modern user interface for technology supervision and control, particularly in the area of clean and waste water distribution systems. The thesis concludes with a description of a user interface based on defined requirements and finally compares them with existing systems.

Nowadays almost all motorized vehicles use electronic control units (ECUs) to control parts of a vehicle’s function. A good way to understand a vehicle’s behaviour is to analyse logging data containing ECU internal variables. Data must then be transferred from the ECU to a computer in order to study such data. Today, Keyword Protocol (KWP) requests are used to read data from the ECUs at Scania. The method is not suitable if many signals should be logged with a higher transfer rate than the one used today. In this thesis, communication protocols, that allow an ECU to communicate with a computer, are studied. The purpose of this master’s thesis is to examine how the transfer rate of variables from Scania’s ECUs to a computer can become faster compared to the method used today in order to get a more frequent logging of the variables. The method that was chosen was implemented, evaluated and also compared to the method used today. The busload, total CPU load and CPU load for the frequency used during the experiments, 100 Hz, was also examined and evaluated. The experiments performed show that the method chosen, data acquisition (DAQ) with CAN Calibration Protocol (CCP), increased the transfer rate of the internal ECU variables significantly compared to the method using KWP requests. The results also show that the number of signals have a major impact on the busload for DAQ. The busload is the parameter that limits the number of signals that can be logged. The total CPU load and the CPU load for 100 Hz are not affected significantly compared to when no transmissions are performed. Even though the busload can become high if many variables are used in DAQ, DAQ with CCP is preferable over KWP requests. This is due to the great increase in transfer rate of the ECU internal variables and thus a great increase in the logging frequency. / Nuförtiden används styrenheter (ECUer) för att styra delar av ett fordons funktion i så gott som alla motoriserade fordon. Ett bra sätt att förstå ett fordons beteende är att analysera loggningsdata som innehåller interna styrenhetsvariabler. Data måste då överföras från styrenheten till en dator för att data ska kunna studeras. Idag används Keyword Protocol-förfrågningar (KWP-förfrågningar) för att läsa data från Scanias styrenheter. Metoden är inte lämplig om man vill logga många variabler med en högre överföringshastighet än den som används idag. I detta examensarbete studeras kommunikationsprotokoll som tillåter en styrenhet att kommunicera med en dator. Examensarbetets syfte är undersöka hur överföringshastigheten av variablerna, från Scanias styrenheter till en dator, kan ökas jämfört med den metod som används idag för att få en mer frekvent loggning av variablerna. Metoden som valdes implementerades, utvärderades och jämfördes med metoden som används idag. Busslasten, totala CPU-lasten och CPU-lasten för den frekvens som användes under experimenten 100 Hz har också undersökts och evaluerats. De utförda experimenten visar att den valda metoden, data acquisition (DAQ) med CAN Calibration Protocol (CCP), ökade överföringshastigheten av de interna styrenhetsvariablerna betydligt jämfört med metoden KWP-förfrågningar. Experimenten visar också att antalet signaler har stor inverkan på busslasten för DAQ. Busslasten är den parameter som begränsar antalet signaler som kan loggas. Den totala CPU-lasten och CPU-lasten för 100 Hz påverkas inte betydligt jämfört med när inga överföringar görs. DAQ med CCP är att föredra framför KWP-förfrågningar även om busslasten blir hög för DAQ då den stora ökningen i överföringshastighet av de interna styrenhetsvariablerna medför en mer frekvent loggning av variablerna.

Industrial Ethernet Communication protocols for Real-Time communication is the focus of research in a variety of different industrial Applications. The subject of thesis study is a comprehensive analysis of Industrial Ethernet protocols used in safety-critical real-time environment, comparing these based on performance, safety, diagnostics and error detection features, topology preference, cost incurred and openness of the protocol. The protocols chosen for this comparisons are ModbusTCP, EtherNet/IP, Ethernet Powerlink, Profinet IO, EtherCAT®, Sercos III, and CC-LINK IE, because they are most relevant for real-time applications and they have a widespread global presence in a lot of industrial applications. The industry in focus here is the spacecraft industry, which is a perfect example of a real-time safety-critical environment, where the protocols need to perform under harsh conditions, such as vibrations, exposure to radiation , extreme temperatures and low atmospheric pressure, and have strict timing requirements. The communication protocols for this environment are held to very high standards of performance and communication accuracy. The challenge here is to implement a protocol that can provide deterministic, real-time communication in such extreme conditions. The protocol in focus here is the ModbusTCP protocol , that is implemented on the ESBO-DS Stratospheric Balloon project to establish a communication link with the pointing system on board. ModbusTCP is chosen for this application because it is reliable, has a large support base, it is easy to implement, transparent and extremely interoperable. All these benefits along with providing real-time communication on a space-based application gives us a good reason to continuing the use of ModbusTCP here. The conclusion of the comparison of the selected real-time communication Industrial Ethernet protocols is that there is no “best ” protocol, and the choice of protocol depends on the application in focus. EtherCAT, SERCOS III and PROFINET IRT are all powerful hard real-time protocols, but require dedicated and expensive hardware for implementation. CC-Link IE is also a popular hard real-time protocol that uses special hardware and is well-established in the East Asian countries. Ethernet/IP and ModbusTCP control a large market share in the Industrial Communication domain in addition to providing good performance easily available tools. Ethernet Powerlink is open and provides both implementation and specification royalty free on its organisations website. / Industriella Ethernet-kommunikationsprotokoll för realtidskommunikation är fokus för forskning inom en mängd olika industriella tillämpningar. Ämnet för avhandlingsstudien är en omfattande analys av industriella Ethernet-protokoll som används i säkerhetskritiska realtidsmiljöer, som jämför dessa baserat på prestanda, säkerhet, diagnostik och feldetekteringsfunktioner, topologipreferenser, uppkomna kostnader och protokollets öppenhet. De valda protokollen för dessa jämförelser är ModbusTCP, EtherNet/IP, Ethernet Powerlink, Profinet IO, EtherCAT®, Sercos III och CC-LINK IE, eftersom de är mest relevanta för realtidsapplikationer och de har en utbredd global närvaro i många industriella tillämpningar. Branschen i fokus här är rymdfarkostindustrin, som är ett perfekt exempel på en realtidssäkerhetskritisk miljö, där protokollen måste fungera under tuffa förhållanden, såsom vibrationer, exponering för strålning, extrema temperaturer och lågt atmosfärstryck, och har strikta tidskrav. Kommunikationsprotokollen för denna miljö håller mycket höga standarder för prestanda och kommunikationsnoggrannhet. Utmaningen här är att implementera ett protokoll som kan tillhandahålla deterministisk kommunikation i realtid under sådana extrema förhållanden. Protokollet i fokus här är ModbusTCP-protokollet, som är implementerat i ESBO-DS Stratospheric Balloon-projektet för att upprätta en kommunikationslänk med peksystemet ombord. ModbusTCP är vald för denna applikation eftersom den är pålitlig, har en stor supportbas, den är lätt att implementera, transparent och extremt driftskompatibel. Alla dessa fördelar tillsammans med att tillhandahålla realtidskommunikation på en rymdbaserad applikation ger oss en god anledning att fortsätta använda ModbusTCP här. Slutsatsen av jämförelsen av de valda realtidskommunikationens Industrial Ethernet-protokoll är att det inte finns något bästaprotokoll, och valet av protokoll beror på applikationen i fokus. EtherCAT, SERCOS III och PROFINET IRT är alla kraftfulla hårda realtidsprotokoll, men kräver dedikerad och dyr hårdvara för implementering. CC-Link IE är också ett populärt hårt realtidsprotokoll som använder speciell hårdvara och är väletablerat i de östasiatiska länderna. Ethernet/IP och ModbusTCP kontrollerar en stor marknadsandel inom industriell kommunikationsdomän förutom att de tillhandahåller bra prestanda lättillgängliga verktyg. Ethernet Powerlink är öppet och tillhandahåller både implementering och specifikation royaltyfritt på sin organisations webbplats.

<p>Ova doktorska disertacija predstavlja istraživački rad sproveden sa ciljem da re&scaron;i problem koji se odnosi na realizaciju povezivanja laboratorijskog, medicinskog mernog uređaja sa personalnim računarom. Rezultat istraživačkog rada je op&scaron;ta metoda koja obezbeđuje efikasnu konverziju podataka I2S protokola u TCP/IP protokol.</p> / <p>This doctoral thesis presents the research work carried out in order to solve a practical problem, the realization of a connection between a medical measuring device and a personal computer. The result of the research is a general method that provides efficient data conversion from the I2S protocol to the TCP/IP protocol.</p>

We are concerned with addressing the worst-case distributed source coding (DSC) problem in asymmetric and interactive communication scenarios and its application to data-gathering wireless sensor networks in enhancing their lifetime. First, we propose a unified canonical framework, obtained by considering different communication constraints and objectives, to address the variants of DSC problem. Second, as for the worst-case information-theoretic analysis, the notion of information entropy cannot be used, we propose information ambiguity, derive its various properties, and prove that it is a valid information measure. Third, for a few variants of our interest of DSC problem, we provide the communication protocols and prove their optimality. In a typical data-gathering sensor network, the base-station that wants to gather sensor data is often assumed to be much more resourceful with respect to energy, computation, and communication capabilities compared to sensor nodes. Therefore, we argue that in such networks, the base-station should bear the most of the burden of communication and computation in the network. Allowing the base-station and sensor nodes to interactively communicate with each other enables us to carry this out. Our definition of sensor network lifetime allows us to reduce the problem of maximizing the worst-case network lifetime to the problem of minimizing the number of bits communicated by the nodes in the worst-case, which is further reduced to the worst-case DSC problem in asymmetric and interactive communication scenarios, with the assumption that the base-station knows the support-set of sensor data. We demonstrate that the optimal solutions of the energy-oblivious DSC problem variants cannot be directly applied to the data-gathering sensor networks, as those may be inefficient in the energy-constrained sensor networks. We address a few energy-efficient variants of DSC problem and provide optimal communication protocols for the sensor networks, based on those variants. Finally, we combine distributed source coding with two other system level opportunities of channel coding and cooperative nature of the nodes to further enhance the lifetime of the sensor networks. We address various scenarios and demonstrate the dependence of the computational complexity of the network lifetime maximization problem on the complex interplay of above system-level opportunities.

QoS routing is one of the major building blocks for supporting QoS in communication networks and, hence, a necessary component of future communication networks. Bandwidth- Constrained Routing Algorithm (BCRA) may help to satisfy QoS requirements such as end-to-end delay, delay-jitter etc when WFQ-like (Weighted Fair Queuing) scheduling mechanisms are deployed. The existing algorithms for bandwidth constrained routing suffer from high message overhead and have a high computational and space complexity. The work presented in the thesis, therefore, focuses on the different techniques that an be used to reserve bandwidth for a unicast connection with low protocol overhead in terms of number of messages. We have compared the performance of the proposed routing algorithms using simulation studies with other bandwidth constrained routing algorithms. The call blocking ratio and message overhead have been used as the performance metric to compare the proposed algorithm with the existing ones. We present three source routing algorithms for unicast connections satisfying the band- width requirement. The first two routing algorithms are based on the partitioning of the network. The link-state broadcasts are limited to the partition. In the first algorithm, the source node queries the other partitions for the state information on a connection request and computes the path based on the information received from the other partitions. The second algorithm is based on state aggregation. The aggregated state of other partitions is maintained at every node. The source node finds a feasible path based on the aggregated information. The path is expanded in every partition, if required, at the time of resource reservation. The third QoS routing algorithm uses the Distance Vector Tables to find a route for a connection. If the shortest path satisfies the bandwidth requirement, then it is selected; otherwise a random deviation is taken at the point where bandwidth requirement is not satisfied and shortest path algorithm is again followed. In all the three algorithms presented, the packets carry the entire path information to the destination node. Therefore, no per connection information is required to be maintained at the intermediate nodes. Simulation results indicate that the proposed algorithms indeed help educing the protocol overhead considerably, and at the same time they give comparable or better performance in terms of resource utilization across a wide range of workloads.

An ad hoc wireless network is an infrastructure-less, autonomous system of nodes connected through wireless links. In many current applications of ad hoc wireless networks, individual wireless nodes are autonomous, rational, and intelligent and are often referred to as selfish nodes, following game theoretic terminology. In an ad hoc wireless network, a typical node may be an intermediate node of a route from a source node to a destination node and therefore is often required to forward packets so as to enable communication to be established. Selfish nodes may not always forward the packets since the forwarding activity consumes the node’s own resources. Such behavior by individual nodes may lead to suboptimal situations where nodes, through their actions, lead to a state that is undesirable from an overall network viewpoint. To counter this, there is a need to stimulate cooperation through methods such as providing appropriate incentives. In this thesis, our interest is in designing rigorous incentive based methods for stimulating cooperation among wireless nodes, in the specific context of broadcast. In particular, we address the Incentive Compatible Broadcast problem: how do we design broadcast protocols that induce truth revelation by the individual wireless nodes? We do this using a game theory and mechanism design framework. Incentive compatibility of broadcast protocols could manifest in two forms: (1) Dominant Strategy Incentive Compatibility (DSIC) (also called strategy-proofness) and (2) Bayesian incentive compatibility (BIC). A DSIC broadcast protocol is one which makes it a best response for every wireless node to reveal its true type, regardless of what the other nodes reveal. A BIC broadcast protocol is one which makes truth revelation a best response for a node, given that the other nodes are truthful. The DSIC property is stronger and more desirable but more difficult to achieve. On the other hand, the BIC property is much weaker and easier to achieve. In this thesis, we first design a DSIC broadcast protocol for ad hoc networks using the well known VCG (Vickrey-Clarke-Groves) mechanisms and investigate its properties and performance. Next, we design a BIC broadcast protocol, investigate its properties, and compare its performance with that of the DSIC broadcast protocol. Both the protocols developed in this thesis provide an elegant solution to the incentive compatible broadcast problem in ad hoc networks with selfish nodes and help stimulate cooperation among the selfish wireless nodes.