The wireless sensor network is a key technology in the 21st century because it has multitude applications and it becomes the new way of interaction between physical environment and computer system. Moreover, the wireless sensor network is a high resource constraint system. Consequently, the techniques used for the development of traditional embedded systems cannot be directly applied. Today wireless sensor nodes were implemented by using only one single processor architecture. This approach does not achieve a robust and efficient energy wireless sensor network for applications such as precision agriculture (outdoor) and telemedicine. The aim of this thesis is to develop a new approach for the realization of a wireless sensor network node using multicore architecture to enable to increase both its robustness and lifetime (reduce energy consumption).

Wireline telecommunications infrastructure in the customer access network or CAN is undergoing a veritable technological and commercial revolution. The paired-copper CAN is being modernised with optical fibre deployed ever closer to customers, culminating soon with fibre-to-the-home networks or some variant thereof. Although bandwidth ceases to be a scarce commodity, the underlying natural monopoly will most likely be strengthened. National competition policy desires open access to multiple service providers yet commercial pressure calls for closure. This has been the recent experience with the hybrid fibre coaxial networks delivering pay television and Internet access. This research asks the question: What are the factors that prevent open access to the broadband services of next generation wireline infrastructure? How can these obstacles be overcome? A particular focus is given to non-price considerations which come to the fore due to the unique strategic and technological characteristics of optical fibre in the access network. The methodological approach involves data gathering via three case studies - that of the Telstra/Foxtel pay television network, the TransACT broadband network and fibre-to-the-home networks in general. Although the ultimate focus is on the research question above, these cases are discussed in a holistic way with consideration of a number of contextual factors. The research also examines the relationship between the concepts of 'open access' and 'network neutrality', visiting the concept of 'common carriage' in doing so. Several findings are reached that illuminate the field of telecommunications access regulation as applied to infrastructure capable of delivering truly next generation broadband services. Since 1993, our politicians have only paid lip service to the importance of competition and have deferred to the demands of the dominant builder of telecommunications infrastructure. From the viewpoints of end-users and access seekers, the access regime is found to be incapable of dealing with the technical and commercial bottlenecks arising from optical fibre in the CAN. It is concluded that communication between users should be recognised as the prime purpose of telecommunications and that the regulatory regime should not reward discriminatory practices detracting from the development of a networked information economy. It is also concluded that dominant players should never be rewarded with access holidays which could otherwise entrench market dominance through the creation of new bottlenecks. Access regulation is ill-equipped to cope with optical fibre in the CAN until it also recognizes the strategic potential of such infrastructure.

Τα τελευταια χρόνια ειμαστε μαρτυρες ενός εντυπωσιακού παραδόξου στο χώρο της τεχνολογίας. Ενώ όλη η ανθρωπότητα αναζητά το κάτι παραπάνω σε υπολογιστική ισχύ, σε χώρο αποθήκευσης και σε ταχύτητα, μικρές αυτόνομες συσκευές έρχονται να κατακτήσουν όλο και περισσότερους τομείς της καθημερινότητάς μας. Μονάδες με την ικανότητα της επεξεργασίας και της αποθήκευσης δεδομένων, της αίσθησης του περιβάλλοντος αλλά και της επικοινωνίας μεταξύ τους, ενσωματώνονται σχεδόν σε όλους τους τομείς δραστηριοποίησης του ανθρώπου. Ο τρόπος αυτό-οργάνωσης αυτών των κόμβων - μονάδων στα πλαίσια μεγάλων ασύρματων δικτύων αισθητήρων, και η συλλογή δεδομένων από μια ευρεία περιοχή, τους δίνει τη δυνατότητα λήψης αποφάσεων, πολλές φορές κρίσιμων, ανάλογα με τις επικρατούμενες συνθήκες. Τα παραπάνω χαρακτηριστικά των δομικών στοιχείων των ασυρμάτων δικτύων αισθητήρων, σε συνδυασμό με τις υπηρεσίες της υψηλής και καθιερωμένης τεχνολογίας του συστήματος GPS (Global Positioning System) μπορούν να δημιουργήσουν έναν εκρηκτικό συνδυασμό και ερευνητικές ιδέες για ανάπτυξη αστικών εφαρμογών αυτοματισμού, στο πλαίσιο των λεγόμενων «έξυπνων πόλεων». Μέσα σε ένα απέραντο δίκτυο ασυρμάτων αισθητήρων, αποτελούμενο από χιλιάδες οντότητες, η ικανότητα του κάθε κόμβου να εντοπίζει την θέση του και να την μοιράζεται με τους υπόλοιπους, κερδίζει ολοένα και περισσότερο έδαφος ειδικά σε θέματα δρομολόγησης. Η χρήση της τεχνολογίας GPS από έναν κόμβο, αντικείμενο που μελετάμε στη παρούσα εργασία, μπορεί να οδηγήσει από την επιτυχή εφαρμογή του Geographical Routing σε ένα WSN, μέχρι την υλοποίηση απαιτητικών εφαρμογών για αντιγραφή κίνησης αλλά και δημιουργίας «έξυπνων αυτοκινητόδρομων» μέσα στους οποίους θα γίνεται αυτόματη πλοήγηση των κινητών υπό την επίβλεψη ενός μεγάλου WSN με κόμβους που συνεργάζονται με το σύστημα δορυφορικού εντοπισμού θέσης αλλά και μεταξύ τους. Όπως γίνεται άμεσα αντιληπτό, η παρούσα εργασία, που πραγματεύεται την σωστή συνεργασία και επικοινωνία ενός δικτύου ασυρμάτων αισθητήρων με την τεχνολογία GPS για την ιχνηλάτηση της πορείας ενός κινητού, μπορεί να δώσει μελλοντική τροφή για ενασχόληση σε πολλαπλά ερευνητικά επίπεδα. Η εργασία μας χωρίστηκε σε τρείς θεματικές ενότητες. Η πρώτη αφορά τον τομέα του WSN και του προγραμματισμού των κόμβων, για επικοινωνία με τους δορυφόρους, με τον υπολογιστή αλλά και μεταξύ τους, με τις απαραίτητες λειτουργικές εφαρμογές που κληθήκαμε να υλοποιήσουμε. Η δεύτερη αναφέρεται στην εργασία μας από την πλευρά του υπολογιστή, που λειτουργεί σαν συλλέκτης των δεδομένων του δικτύου, και τη σύνδεση του συστήματός μας με την πρότυπη εφαρμογή Google EarthTM για απεικόνιση σε πραγματικό χρόνο της τροχιάς που ακολουθεί ένα κινητό. Η τρίτη ενότητα περιλαμβάνει εκτέλεση πειραμάτων χρήσης του συστήματος που σχεδιάσαμε και παράθεση οπτικοποιημένων αποτελεσμάτων, για την εύκολη εξαγωγή χρήσιμων συμπερασμάτων όσον αφορά στη λειτουργικότητά του. / The last few years we witness a striking paradox in the field of technology. While all humanity is seeking for more computing power, more storage capabilities and more proccessing speed, small autonomous devices have appeared to occupy more and more areas of our lives in the daily routine. Units with the capabilities of data proccessing, data storage, enviromental sensing and communication with each other, are incorporated in almost all areas of human activity. The way in which these nodes are self-organized into large Wireless Sensor Networks and their ability to collect data from a wide area, enables them in taking decisions, critical ones sometimes, according to the prevailing conditions. The above characteristics of the components of wireless sensor networks, combined with the services of the established and standard technology of the GPS (Global Positioning System) can create ideas for research and development of urban applications of automation in the so-called “smart cities”. In a vast network of wireless sensors, consisting of thousands of entities, the ability of each node to detect its position and share it with others, is gaining more and more ground especially in routing topics. The use of GPS technology from a node, which is one of the subjects we study in this thesis, could lead from the successful implementation of the Geographical Routing, to the implementation of more demanding applications for motion replication but also to creating “smart highways” through which mobiles will execute an automatic navigation, supervised by a large WSN consisting of nodes cooperating with the satellite positioning system and with each other. As is readily apparent, the present thesis which is dealing with the good cooperation and communication of a wireless sensor network with the GPS technology in order to trace the path of a mobile, can provide room for future involvement in multiple levels of research. Our work was divided into three thematic sections. The first concerns the field of WSN and node programming so that they can be able to communicate with satellites, with the computer and with each other, by installing them with the necessary functional applications that we had to implement. The second section refers to our work on the computer side, which acts as a network data collector, and the link of our system with the standard Google EarthTM application to display the path of a mobile in real – time. The third section includes conducting experiments using the system we designed. The results are visualised and listed for an easy extraction of usefull conclusions regarding the functionality of our system.

Network management plays a vital role to keep a network and its application work e ciently. The network management in MANETs is a crucial and the challenging task, as these networks are characterized by dynamic environment and the scarcity of resources. There are various existing approaches for network management in MANETs. The Ad hoc Network Management Protocol (ANMP) has been one of the rst e orts and introduced an SNMP-based solution for MANETs. An alternative SNMP-based solu-tion is proposed by GUERRILLA Management Architecture (GMA). Due to self-organizing characteristic feature of MANETs, the management task has to be distributed. Policy-based network management relatively o ers this feature, by executing and applying policies pre-viously de ned by network manager. Otherwise, the complexity of realization and control becomes di cult Most of the works address the current status of the MANET to take the network man-agement decisions. Currently, MANETs addresses the dynamic and intelligent decisions by considering the present situation and all related history information of nodes into consid-eration. In this connection we have proposed a network management system using agents (NMSA) for MANETs, resolving major issues like, node monitoring, location management, resource management and QoS management. Solutions to these issues are discussed as inde-pendent protocols, and are nally combined into a single network management system, i.e., NMSA. Agents are autonomous, problem-solving computational entities capable of performing e ective operation in dynamic environments. Agents have cooperation, intelligence, and mobility characteristics as advantages. The agent platforms provide the di erent services to agents, like execution, mobility, communication, security, tracking, persistence and directory etc. The platform execution environment allows the agents to run, and mobility service allows them to travel among the di erent execution environments. The entire management task will be delegated to agents, which then executes the management logic in a distributed and autonomous fashion. In our work we used the static and mobile agents to nd some solutions to the management issues in a MANET. We have proposed a node monitoring protocol for MANETs, which uses both static agent (SA) and mobile agents (MA), to monitor the nodes status in the network. It monitors the gradational energy loss, bu er, bandwidth, and the mobility of nodes running with low to high load of mobile applications. Protocol assumes the MANET is divided into zones and sectors. The functioning of the protocol is divided into two segments, The NMP main segment, which runs at the chosen resource rich node (RRN) at the center of a MANET, makes use of SA which resides at same RRN, and the NMP subsegment which runs in the migrated MAs at the other nodes. Initially SA creates MAs and dispatches one MA to each zone, in order to monitor health conditions and mobility of nodes of the network. MAs carrying NMP subsegment migrates into the sector of a respective zone, and monitors the resources such as bandwidth, bu er, energy level and mobility of nodes. After collecting the nodes information and before moving to next sector they transfer collected information to SA respectively. SA in turn coordinates with other modules to analyze the nodes status information. We have validated the protocol by performing the conformance testing of the proposed node monitoring protocol (NMP) for MANETs. We used SDL to obtain MSCs, that repre-sents the scenario descriptions by sequence diagrams, which in turn generate test cases and test sequences. Then TTCN-3 is used to execute the test cases with respect to generated test sequences to know the conformance of protocol against the given speci cation. We have proposed a location management protocol for locating the nodes of a MANET, to maintain uninterrupted high-quality service for distributed applications by intelligently anticipating the change of location of its nodes by chosen neighborhood nodes. The LMP main segment of the protocol, which runs at the chosen RRN located at the center of a MANET, uses SA to coordinate with other modules and MA to predict the nodes with abrupt movement, and does the replacement with the chosen nodes nearby which have less mobility. We have proposed a resource management protocol for MANETs, The protocol makes use of SA and MA for fair allocation of resources among the nodes of a MANET. The RMP main segment of the protocol, which runs at the chosen RRN located at the center of a MANET, uses SA to coordinate with other modules and MA to allocate the resources among the nodes running di erent applications based on priority. The protocol does the distribution and parallelism of message propagation (mobile agent with information) in an e cient way in order to minimize the number of message passing with reduction in usage of network resources and improving the scalability of the network. We have proposed a QoS management protocol for MANETs, The QMP main segment of the protocol, which runs at the chosen RRN located at the center of a MANET, uses SA to coordinate with other modules and MA to allocate the resources among the nodes running di erent applications based on priority over QoS. Later, to reallocate the resources among the priority applications based on negotiation and renegotiation for varying QoS requirements. The performance testing of the protocol is carried out using TTCN-3. The generated test cases for the de ned QoS requirements are executed with TTCN-3, for testing of the associated QoS parameters, which leads to performance testing of proposed QoS management protocol for MANETs. We have combined the developed independent protocols for node monitoring, location management, resource management, and QoS management, into one single network management system called Network Management System using Agents (NMSA) for MANETs and tested in di erent environments. We have implemented NMSA on Java Agent development environment (JADE) Platform. Our developed network management system is a distributed system. It is basically divided into two parts, the Network Management Main Segment and other is Network Management Subsegment. A resource rich node (RRN) which is chosen at the center of a MANET where the Main segment of NMSA is located, and it controls the management activities. The other mobile nodes in the network will run MA which has the subsegments of NMSA. The network management system, i.e., the developed NMSA, has Network manage-ment main (NMSA main), Zones and sector segregation scheme, NMP, LMP, RMP, QMP main segments at the RRN along with SA deployed. The migrated MA at mobile node has subsegments of NMP, LMP, RMP, and QMP respectively. NMSA uses two databases, namely, Zones and sectors database and Node history database. Implementation of the proposed work is carried out in a con ned environment with, JDK and JADE installed on network nodes. The launched platform will have AMS and DF automatically generated along with MTP for exchange of message over the channel. Since only one JVM, which is installed, will executes on many hosts in order to provide the containers for agents on those hosts. It is the environment which o ered, for execution of agents. Many agents can be executed in parallel. The main container, is the one which has AMS and DF, and RMI registry are part of JADE environment which o ers complete run time environment for execution of agents. The distribution of the platform on many containers of nodes is shown in Fig. 1. The NMSA is based on Linux platform which provides distributed environment, and the container of JADE could run on various platforms. JAVA is the language used for code development. A middle layer, i.e., JDBC (java database connection) with SQL provides connectivity to the database and the application. The results of experiments suggest that the proposed protocols are e ective and will bring, dynamism and adaptiveness to the applied system and also reduction in terms network overhead (less bandwidth consumption) and response time.

In the introduction there is described a history of the post office. There is also outlined a current state and a future intention of Czech Post (Česká pošta s. p.). Further there is explained an issue of the formatting of the address side of postcards and writing. In another part there is a job description of the sorting machine SIEMENS IRV 3000 and also information on the location of this sorting machine at each collecting transport nodes. In the conclusion there is described a consideration to increase a number of appropriate mail pieces for the sorting machine.

1. Main Contributions made by the supplicant: This thesis proposes an Open Computing Language (OpenCL) framework to address the challenges of implementation of graph algorithms on parallel architectures and large scale graph processing. The proposed framework uses the front-end of the existing Falcon DSL compiler, andso, programmers enjoy conventional, imperative and shared memory programming style. The back-end of the framework generates implementations of graph algorithms in OpenCL to target single device architectures. The generated OpenCL code is portable across various platforms, e.g., CPU and GPU, and also vendors, e.g., NVIDIA, Intel and AMD. The framework automatically generates code for thread management and memory management for the devices. It hides all the lower level programming details from the programmers. A few optimizations are applied to reduce the execution time. The large graph processing challenge is tackled through graph partitioning over multiple devices of a single node and multiple nodes of a distributed cluster. The programmer codes a graph algorithm in Falcon assuming that the graph fits into single machine memory and the framework handles graph partitioning without any intervention by the programmer. The framework analyses the Abstract Syntax Tree (AST) generated by Falcon to find all the necessary information about communication and synchronization. It automatically generates code for message passing to hide the complexity of programming in a distributed environment. The framework also applies a set of optimizations to minimize the communication latency. The thesis reports results of several experiments conducted on widely used graph algorithms: single source shortest path, pagerank and minimum spanning tree to name a few. Experimental evaluations show that the reported results are comparable to the state-of-art non-portable graph DSLs and frameworks on a single node. Experiments in a distributed environment to show the scalability and efficiency of the framework are also described. 2. Summary of the Referees' Written Comments: Extracts from the referees' reports are provided below. A copy of the written replies to the clarifications sought by the external examiner is appended to this report. Referee 1: This thesis extends the Falcon framework with OpenCL for parallel graph processing on multi-device and multi-node architectures. The thesis makes important contributions. Processing large graphs in short time is very important, and making use of multiple nodes and devices is perhaps the only way to achieve this. Towards this, the thesis makes good contributions for easy programming, compiler transformations and efficient runtime systems. One of the commendable aspects of the thesis that it demonstrates with graphs that cannot be accommodated In the memory of a single device. The thesis is generally written well. The related work coverage is very good. The magnitude of thesis excellent for a Masters work. The experimental setup is very comprehensive with good set of graphs, good experimental comparisons with state-of-art works and good platforms. Particularly. the demonstration with a GPU cluster with multiple GPU nodes (Chapter 5) is excellent. The attempt to demonstrate scalability with 2, 4 and 8 nodes is also noteworthy. However, the contributions on optimizations are weak. Most of the optimizations and compiler transformations are straight-forward. There should be summary observations on the results in Chapter 3, especially given that the results are mixed and don't quite clearly convey the clear advantages of their work. The same is the case with multi-device results in chapter 4, where the results are once again mixed. Similarly, the speedups and scalability achieved with multiple nodes are not great. The problem size justification in the multi-node results is not clear. (Referee 1 also indicates a couple of minor changes to the thesis). Referee 2: The thesis uses the OpenCL framework to address the problem of programming graph algorithms on distributed systems. The use of OpenCL ensures that the generated code is platform-agnoistic and vendor-agnoistic. Sufficient experimentation with large scale graphs and reasonable size clusters have been conducted to demonstrate the scalability and portability of the code generated by the framework. The automatically generated code is almost as efficient as manually written code. The thesis is well written and is of high quality. The related work section is well organized and displays a good knowledge of the subject matter under consideration. The author has made important contributions to a good publication as well. 3. An Account of the Open Oral Examination: The oral examination of Ms. Parita Patel took place during 10 AM and 11AM on 27th November 2017, in the Seminar Hall of the Department of Computer Science and Automation. The members of the Oral Examination Board present were, Prof. Sathish Vadhiyar, external examiner and Prof. Y. N. Srikant, research supervisor. The candidate presented the work in an open defense seminar highlighting the problem domain, the methodology used, the investigations carried out by her, and the resulting contributions documented in the thesis before an audience consisting of the examiners, some faculty members, and students. Some of the questions posed by the examiners and the members of the audience during the oral examination are listed below. 1. How much is the overlap between Falcon work and this thesis? Response: We have used the Falcon front end in our work. Further, the existing Falcon compiler was useful to us to test our own implementation of algorithms in Falcon. 2. Why are speedup and scalability not very high with multiple nodes? Response: For the multi-node architecture, we were not able to achieve linear scalability because, with the increase in number of nodes, communication cost increases significantly. Unless the computation cost in the nodes is significant and is much more than the communication cost, this is bound to happen. 3. Do you have plans of making the code available for use by the community? Response: The code includes some part of Falcon implementation (front-end parsing/grammar) also. After discussion with the author of Falcon, the code can be made available to the community. 4. How can a graph that does not fit into a single device fit into a single node in the case of multiple nodes? Response: Single node machine used in the experiments of “multi-device architecture” contains multiple devices while each node used in experiments of “multi-node architecture” contains only a single device. So, the graph which does not fit into single-node-single-device memory can fit into single-node-multi-device after partitioning. 5. Is there a way to permit morph algorithms to be coded in your framework? Response: Currently, our framework does not translate morph algorithms. Supporting morph algorithms will require some kind of runtime system to manage memory on GPU since morph algorithms add and remove the vertices and edges to the graph dynamically. This can be further explored in future work. 6. Is it possible to accommodate FPGA devices in your framework? Response: Yes, we can support FPGA devices (or any other device that is compatible for OpenCL) just by specifying the device type in the command line argument. We did not work with other devices because CPU and GPU are generally used to process graph algorithms. The candidate provided satisfactory answers to all the questions posed and the clarifications sought by the audience and the examiners during the presentation. The candidate's overall performance during the open defense and the oral examination was very satisfactory to the oral examination board. 4. Certificate of Corrections and Changes: All the necessary corrections and changes suggested by the examiners have been made in the thesis and these have been verified by the members of the oral examination board. The thesis has been recommended for acceptance in its revised form. 5. Final Recommendation: In view of the recommendations of the referees and the satisfactory performance of the candidate in the oral examination, the oral examination board recommends that the thesis of Ms. ParitaPatel be accepted for the award of the M.Sc(Engg.) Degree of the Institute. Response to the comments by the external examiner on the M.Sc(Engg.) thesis “Compilation of Graph Algorithms for Hybrid, Cross-Platform, and Distributed Architectures” by Parita Patel 1. Comment: The contributions on optimizations are weak. Response: The novelty of this thesis is to make the Falcon platform agnostic, and additionally process large scale graphs on multi-devices of a single node and multi-node clusters seamlessly. Our framework performs similar to the existing frameworks, but at the same time, it targets several types of architectures which are not possible in the existing works. Advanced optimizations are beyond the scope of this thesis. 2. Comment: The translation of Falcon to OpenCL is simple. While the translation of Falcon to OpenCL was not hard, figuring out the details of the translation for multi-device and multi-node architectures was not simple. For example, design of implementations for collection, set, global variables, concurrency, etc., were non-trivial. These designs have already been explained in the appropriate places in the thesis. Further, such large software introduced its own intricacies during development. 3. Comment: Lines between Falcon work and this work are not clear. Response: Appendix-A shows the falcon implementation of all the algorithms which we used to run the experiments. We compiled these falcon implementations through our framework and subsequently ran the generated code on different types of target architectures and compared the results with other framework's generated code. These falcon programs were written by us. We have also used the front-end of the Falcon compiler and this has already been stated in the thesis (page 16). 4. Comment: There should be a summary of observations in chapter 3. Response: Summary of observations have been added to chapter 3 (pages 35-36), chapter 4 (page 46), and chapter 5 (page 51) of the thesis. 5. Comment: Speedup and scalability achieved with multiple nodes are not great. Response: For the multi-node architecture, we were not able to achieve linear scalability because, with the increase in number of nodes, communication cost increases significantly. Unless the computation cost in the nodes is significant and is much more than the communication cost, this is bound to happen. 6. Comment: It will be good to separate the related work coverage into a separate chapter. Response: The related work is coherent with the flow in chapter 1. It consists of just 4.5 pages and separating it into a separate chapter would make both (rest of) chapter 1 and the new chapter very small. Therefore, we do not recommend it. 7. Comment: The code should be made available for use by the community. Response: The code includes some part of Falcon code (front-end parsing/grammar) also. After discussion with the author of Falcon, the code can be made available to the community. 8. Comment: Page 28: Shouldn’t the else part be inside the kernel? Response: There was some missing text and a few minor changes in Figure 3.14 (page 28) which have been incorporated in the corrected thesis. 9. Comment: Figure 4.1 needs to be explained better. Response: Explanation for Figure 4.1 (pages 38-39) has been added to the thesis. 10. Comment: The problem size justification in the multi-node results is not clear. Response: Single node machine used in the experiments of “multi-device architecture” contains multiple devices while each node used in experiments of “multi-node architecture” contains only a single device. So, the graph which does not fit into single-node-single-device memory can fit into single-node-multi-device after partitioning. Name of the Candidate: Parita Patel (S.R. No. 04-04-00-10-21-14-1-11610) Degree Registered: M.Sc(Engg.) Department: Computer Science & Automation Title of the Thesis: Compilation of Graph Algorithms for Hybrid, Cross-Platform and Graph algorithms are abundantly used in various disciplines. These algorithms perform poorly due to random memory access and negligible spatial locality. In order to improve performance, parallelism exhibited by these algorithms can be exploited by leveraging modern high performance parallel computing resources. Implementing graph algorithms for these parallel architectures requires manual thread management and memory management which becomes tedious for a programmer. Large scale graphs cannot fit into the memory of a single machine. One solution is to partition the graph either on multiple devices of a single node or on multiple nodes of a distributed network. All the available frameworks for such architectures demand unconventional programming which is difficult and error prone. To address these challenges, we propose a framework for compilation of graph algorithms written in an intuitive graph domain-specific language, Falcon. The framework targets shared memory parallel architectures, computational accelerators and distributed architectures (CPU and GPU cluster). First, it analyses the abstract syntax tree (generated by Falcon) and gathers essential information. Subsequently, it generates optimized code in OpenCL for shared-memory parallel architectures and computational accelerators, and OpenCL coupled with MPI code for distributed architectures. Motivation behind generating OpenCL code is its platform-agnostic and vendor-agnostic behavior, i.e., it is portable to all kinds of devices. Our framework makes memory management, thread management, message passing, etc., transparent to the user. None of the available domain-specific languages, frameworks or parallel libraries handle portable implementations of graph algorithms. Experimental evaluations demonstrate that the generated code performs comparably to the state-of-the-art non-portable implementations and hand-tuned implementations. The results also show portability and scalability of our framework.

Parmi les architectures candidates pour les générations sub-22nm figurent les transistors sur silicium sur isolant (SOI). A cette échelle, les composants doivent intégrer des films isolants enterrés (BOX) et des canaux de conduction (Body) ultra-minces. A ceci s'ajoute l'utilisation d'empilements de grille avancés (diélectriques à haute permittivité / métal de grille) et une ingénierie de la contrainte mécanique avec l'utilisation d'alliages SiGe pour le canal des transistors de type P. La mise au point d'une telle technologie demande qu'on soit capable d'extraire de façon non destructive et avec précision la qualité du transport électronique et des interfaces, ainsi que les valeurs des paramètres physiques (dimensions et dopages), qui sont obtenues effectivement en fin de fabrication. Des techniques d'extraction de paramètres ont été développées au cours du temps. L'objectif de cette thèse est de reconsidérer et de faire évoluer ces techniques pour les adapter aux épaisseurs extrêmement réduites des composants étudiés. Elle combine mesures approfondies et modélisation en support. Parmi les résultats originaux obtenus au cours de cette thèse, citons notamment l'adaptation de la méthode split CV complète qui permet désormais d'extraire les paramètres caractérisant l'ensemble de l'empilement SOI, depuis le substrat et son dopage jusqu'à la grille, ainsi qu'une analyse extrêmement détaillée du transport grâce à des mesures en régime de couplage grille arrière à température variable ou l'exploitation de la magnétorésistance de canal depuis le régime linéaire jusqu'en saturation. Le mémoire se termine par une analyse détaillée du bruit basse fréquence. / Silicon on insulator (SOI) transistors are among the best candidates for sub-22nm technology nodes. At this scale, the devices integrate extremely thin buried oxide layers (BOX) and body. They also integrate advanced high-k dielectric / metal gate stacks and strain engineering is used to improve transport properties with, for instance, the use of SiGe alloys in the channel of p-type MOS transistors. The optimization of such a technology requires precise and non-destructive experimental techniques able to provide information about the quality of electron transport and interface quality, as well as about the real values of physical parameters (dimensions and doping level) at the end of the process. Techniques for parameter extraction from electrical characteristics have been developed over time. The aim of this thesis work is to reconsider these methods and to further develop them to account for the extremely small dimensions used for sub-22nm SOI generations. The work is based on extended characterization and modelling in support. Among the original results obtained during this thesis, special notice should be put on the adaptation of the complete split CV method which is now able to extract the characteristic parameters for the entire stack, from the substrate and its doping level to the gate stack, as well as an extremely detailed analysis of electron transport based on low temperature characterization in back-gate electrostatic coupling conditions or the exploitation of channel magnetoresistance from the linear regime of operation to saturation. Finally, a detailed analysis of low-frequency noise closes this study.

Submitted by JÚLIO HEBER SILVA (julioheber@yahoo.com.br) on 2017-09-13T18:01:57Z No. of bitstreams: 2 Dissertação - Roney Lopes Lima - 2017.pdf: 3471034 bytes, checksum: 2dd29fe3cd16f3d5ac0ddabf0ce316b4 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2017-09-19T14:01:45Z (GMT) No. of bitstreams: 2 Dissertação - Roney Lopes Lima - 2017.pdf: 3471034 bytes, checksum: 2dd29fe3cd16f3d5ac0ddabf0ce316b4 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-09-19T14:01:45Z (GMT). No. of bitstreams: 2 Dissertação - Roney Lopes Lima - 2017.pdf: 3471034 bytes, checksum: 2dd29fe3cd16f3d5ac0ddabf0ce316b4 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-08-23 / Fundação de Amparo à Pesquisa do Estado de Goiás - FAPEG / The application of Evolutionary Algorithms in the solution of problems characterized by the unviability through deterministic methods, has made this technique a vast object investigated. Its application to Network Design Problems (NDPs), has been specially studied. NDPs are characterized by modeling real world problems related to network design applied to resource distribution, logistics, telecommunications, routing and even social networks. The solution to these problems involves searching for a graph such as trees that meets criteria for cost minimization, availability, scaling among other constraints that make them complex. The application of Evolutionary Agorithms to NDPs requires a Representation that codes solutions properly towards to these problems. The Node-Depth Encoding (NDE) has been studied and presented results that have aroused the attention of researchers in this topic. In this work, we propose the development of a new recombination operator for NDE called NCX, based on the permutation recombination operator CX. In addition, a method is proposed for correction of infeasible solutions due to an invalid depth for a position in the array. The correction method is applied to both NCX, NOX and NPBX. The operators with their methods of correction are validated for the bias and heritability properties and finally are applied to the Bounded Diameter Minimmum Spanning Tree (BDMSTP) through Evolutionary Algorithms developed for this NDP. The results show that the operators have bias towards to star like trees and good heritability of the edges and depths of the vertices. The developed operators also showed competitiveness when applied to the BDMSTP, even surpassing other representations in the quality of the solutions. / A aplicação de Algoritmos Evolutivos na resolução de problemas caracterizados pela inviabilidade de solução através de métodos determinísticos, fez dessa técnica um objeto vastamente investigado. Sua aplicação para Problemas de Projeto de Redes (PPRs), tem sido especialmente estudada. PPRs são caracterizados por modelar problemas reais relacionados a design de redes aplicados a distribuição de recursos, logística, telecomunicações, roteamento e até mesmo redes sociais. A solução desses problemas envolve a busca de um grafo como uma árvore por exemplo que atenda a critérios de minimização de custos, disponibilidade, escala entre outras restrições que os tornam complexos. A aplicação de Algoritmos Evolutivos a PPRs demanda a utilização de uma Representação que codifique adequadamente soluções para esses problemas. A Representação Nó-Profundidade (RNP) tem sido estudada e apresentado resultados que despertaram a atenção dos pesquisadores nesse tema. Neste trabalho, propõe-se o desenvolvimento de um novo operador de recombinação para a RNP chamado NCX, com base no operador CX de recombinação em permutações. Além disso, é proposto um método para correção de soluções infactíveis devido a profundidade inválida para a posição no \textit{array}. O método de correção é aplicado tanto para NCX, quanto para outros dois operadores de recombinação já desenvolvidos para a RNP, o NOX cujo funcioamento é inspirado no operador OX, e NPBX cujo funcionamento é inspirado no operador PBX. Os operadores com os seus devidos métodos de correção são validados para as propriedades tendência e hereditariedade e por fim são aplicados ao Problema da Árvore Geradora Mínima com Restrição de Diâmetro (BDMSTP) através de Algoritmos Evolutivos desenvolvidos para esse PPR. Os resultados mostram que os operadores possuem tendência para árvores estrela e boa hereditariedade das arestas e das profundidades dos vértices. Os operadores desenvolvidos também mostraram competitividade ao serem aplicados ao BDMSTP, chegando a superar outras representações em qualidade das soluções.

Introdução: Em câncer de mama, existem evidências de que o microambiente pode influenciar o desenvolvimento do tumor no sítio primário, bem como em metástases regionais e a distância. Neste contexto, fibroblastos são importantes células estromais que podem influenciar a proliferação e a migração de células do câncer e podem prover um nicho apropriado para o desenvolvimento tumoral. Objetivos:O principal objetivo deste trabalho é comparar células estromais obtidas do tumor primário (PT), metástase linfonodal (N+) e medula óssea (BM) de pacientes com câncer de mama, através do perfil de expressão gênica. Pacientes e Métodos: Foi analisada a expressão gênica de fibroblastos (cultura primária) de 11 pacientes com câncer de mama. O perfil de expressão foi determinado em PT (n=4), N+(n=3) e BM (n=4) através de uma plataforma de cDNA microarray customizada (contendo 4.800 sequencias imobilizadas, representando cerca de 4600 genes), e os genes diferencialmente expressos foram identificados pelo teste SAM multiclasse, seguido pelo teste SAM de duas classes (TMEV, FDR 0%). A análise funcional foi realizada pelo software DAVID v6.7. Validação técnica foi realizada em 6 amostras previamente analisadas no microarray e a validação biológica em fibroblastos obtidos de outros 16 pacientes utilizando-se de RT-qPCR. Resultados: O perfil de expressão gênica dos fibroblastos obtidos de diferentes sítios mostraram 267 genes diferencialmente expressos, os quais apropriadamente agruparam os fibroblastos de acordo com suas origens (PT vs. N+ vs. BM). Apesar das diferenças entre PT e N+ serem representadas por 20 genes, as diferenças entre PT vs. BM e N+ vs BM foram mais significantes (235 e 245 genes diferencialmente expressos respectivamente). Análise funcional dos genes diferencialmente expressos mostrou enriquecimento de funções relacionadas ao desenvolvimento e morfogênese.A seguir, a expressão de alguns genes selecionados foi analisada em uma série diferente de amostras (validação biológica). Desse modo observamos que NOTCH2 confirmou uma alta expressão em N+ (vs. PT), e ADCY2, HECTD1, HNMT, LOX, MACF1 e USP16 confirmaram alta expressão em BM (vs PT). Conclusão:Em pacientes com câncer de mama, células estromais obtidas de diferentes origens apresentam um perfil de expressão gênica diferencial, o qual pode influenciar o comportamento do tumor / may influence tumor development in the primary site of breast cancer, as well as in regional and distant metastatic sites. In this context, fibroblasts are important stromal cells which influence proliferation and migration of cancer cells and may also provide an appropriate niche to tumor development. Objectives: The main objective of this work is the comparison of stromal cells from the primary tumor (PT), lymph node metastasis (N+) and bone marrow (BM) obtained from breast cancer patients, through gene expression profile. Patients and Methods: The gene expression profile was analyzed in fibroblasts primary culture from 11 breast cancer patients. The expression profiles of PT cells (n=4), N+ cells (n=3) and BM cells (n=4) were determined through a customized cDNA microarray platform (containing 4800 immobilized sequences which represents 4600 genes approximately). The analysis were performed by SAM multiclass (TMEV; FDR 0%), followed by SAM two classes test (TMEV; FDR 0%). Functional analysis was performed using DAVID v6.7. Technical validation was performed in same 6 samples that were previously analyzed in microarray experiments and biological validation was performed in fibroblasts obtained from other group of 16patients by RT-qPCR Results: The expression profile of fibroblasts obtained from three sites revealed 267 differentially expressed genes, which appropriately clustered fibroblasts in three different branches, in accordance with their origin (PT vs. N+ vs. BM). Although the differences between PT and N+ were represented by 20 genes, differences between PT vs. BM and N+ vs. BM were more significant (235 and 245 differentially expressed genes respectively). Functional analysis revealed enrichment of functions related to development and morphogenesis. Afterwards, the expression of some selected genes were analyzed in a different batch of samples (biological validation).Thereby, NOTCH2 confirmed high expression in N+ (vs. PT), and ADCY2, HECTD1, HNMT, LOX, MACF1 and USP16 confirmed high expression in BM (vs. PT). Conclusion: In breast cancer patients, stromal cells obtained from different origins present a differential gene expression profile, which may influence tumor behavior

Orientador: Luiz Carlos Zeferino / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Ciencias Medicas / Made available in DSpace on 2018-08-12T08:55:33Z (GMT). No. of bitstreams: 1 Vieira_SabasCarlos_D.pdf: 1418388 bytes, checksum: 1d70cc451697cfbd93e48319a0284d4e (MD5) Previous issue date: 2008 / Resumo: Objetivos: Avaliar a detecção do linfonodo sentinela em pacientes com câncer do colo do útero utilizando a combinação de azul patente com tecnécio99m e complicações associadas ao uso do azul patente. Sujeitos e métodos: Este foi um estudo de uma série de casos, para o qual foram selecionadas 56 mulheres com diagnóstico de câncer do colo do útero estádios Ia2,Ib1,Ib2 e IIa da FIGO, que se submeteram ao procedimento de identificação do linfonodo sentinela. O período de realização do estudo foi de maio de 2006 a dezembro de 2007. O estudo é apresentado em dois artigos: o primeiro consiste na detecção do linfonodo sentinela no câncer do colo do útero pela combinação do azul patente com tecnécio 99m e avalia a concordância entre a linfocintigrafia pré-operatória e o mapeamento linfático intra-operatório com o gama probe; o segundo consiste na avaliação das alterações da oximetria de pulso das pacientes submetidas à cirurgia após a injeção do azul patente. Resultados: No primeiro artigo identificou-se pelo menos um linfonodo sentinela em 83,13% das pacientes e a localização mais freqüente destes linfonodos foi na cadeia ilíaca externa. A sensibilidade, especificidade, valor preditivo positivo e valor preditivo negativo foram, respectivamente, 80%, 100%, 100% e 97,67% no histopatológico de congelação. Além disso, observou-se que linfocintigrafia pré operatória detecta um número consideravelmente menor de linfonodos sentinelas quando comparado ao mapeamento linfático intra-operatório com o gama probe. No segundo estudo observou-se que somente uma paciente apresentou reação anafilática. Treze pacientes apresentaram queda de oximetria de pulso (menor que 96% de saturação) após a injeção do azul patente no colo do útero, que durou em média cinco minutos e sem repercussões clínicas; essa queda se associou de forma limítrofe com tumores maiores e localizados ao redor do orifício externo do canal cervical. Conclusões: Concluiu-se que a combinação do azul patente com o tecnécio99m demonstrou excelentes resultados na detecção do linfonodo sentinela; a linfocintigrafia pré-operatória não oferece qualquer vantagem em relação ao mapeamento linfático intra-operatório com azul patente e tecnécio99m. Não houve repercussões clínicas devido à queda da oximetria de pulso e essas alterações se correlacionaram, embora com significância limítrofe, com tumores maiores e localizados ao redor do orifício cervical externo. / Abstract: Objectives: To evaluate sentinel lymph node detection in cervical cancer patients using a combination of patent blue dye and technetium99m and assess complications associated with the use of patent blue dye. Subjects and methods: This study investigated a case series that selected 56 women diagnosed with FIGO stage Ia2, Ib1, Ib2 and IIa cervical cancer who underwent a procedure for sentinel lymph node identification. The study was conducted from May 2006 to December 2007 and was described in two articles. The first article was about a study of sentinel lymph node detection in cervical cancer using a combination of patent blue dye and technetium99m. It assessed the agreement between preoperative lymphoscintigraphy and intraoperative lymphatic mapping with a gamma probe. The second article focused on the evaluation of changes in pulse oximetry readings in patients undergoing surgery after patent blue injection. Results: In the first article, at least one sentinel lymph node was identified in 83.13% of the patients and the most frequent site for finding sentinel lymph nodes was the external iliac chain. The sensitivity, specificity, positive predictive value and negative predictive value were 80%, 100%, 100% and 97.67% respectively on histopathology examination of frozen biopsy. In addition, it was observed that preoperative lymphoscintigraphy detected a substantially lower number of sentinel lymph nodes when compared to intraoperative lymphatic mapping with a gamma probe. In the second study, only one patient presented with an anaphylactic reaction. Thirteen patients showed a decrease in pulse oximetry readings (less than 96% saturation) after patent blue injection into the cervix, which lasted an average of five minutes and had no clinical repercussions. There was a borderline association between this decline in oxygen saturation values and tumors that were larger and located around the external cervical os. Conclusions: It was concluded that a combination of patent blue dye and technetium99m demonstrated excellent results in the detection of sentinel lymph nodes. Preoperative lymphoscintigraphy offers no advantage in relation to intraoperative lymphatic mapping with patent blue dye and technetium99m. There were no clinical repercussions due to lower oxygen saturation values. These changes correlated with tumors that were larger and located around the external cervical os, although the significance of this correlation was borderline. / Doutorado / Ciencias Biomedicas / Doutor em Tocoginecologia