Typescript (photocopy). / Digitized by Kansas Correctional Industries

ITC/USA 2008 Conference Proceedings / The Forty-Fourth Annual International Telemetering Conference and Technical Exhibition / October 27-30, 2008 / Town and Country Resort & Convention Center, San Diego, California / This paper describes the design and implementation of a test instrumentation network configuration and verification system. Given a multivendor instrument part catalog that contains sensor, actuator, transducer and other instrument data; user requirements (including desired measurement functions) and technical specifications; the instrumentation network configurator will select and connect instruments from the catalog that meet the requirements and technical specifications. The instrumentation network configurator will enable the goal of mixing and matching hardware from multiple vendors to develop robust solutions and to reduce the total cost of ownership for creating and maintaining test instrumentation networks.

Performance-hungry data center applications demand increasingly higher performance from their storage in addition to larger capacity memory at lower cost. While the existing storage technologies (e.g., HDD and flash-based SSD) are limited in their performance, the most prevalent memory technology (DRAM) is unable to address the capacity and cost requirements of these applications. Emerging byte-addressable, non-volatile memory technologies (such as PCM and RRAM) offer performance within an order of magnitude of DRAM, prompting their inclusion in the processor memory subsystem. Such load/store accessible non-volatile or persistent memory (referred to as NVM or PM) introduces an interesting new tier that bridges the performance gap between DRAM and PM, and serves the role of fast storage or slower memory. However, PM has several implications on system design, both hardware and software: (i) the hardware caching mechanisms, while necessary for acceptable performance, complicate the ordering and durability of stores to PM, (ii) the high performance of PM (compared to NAND) and the fact that it is byte-addressable necessitate rethinking of the system software to manage PM and the interfaces to expose PM to the applications, and (iii) the future memory-based applications that will likely employ systems coupling PM with DRAM (for cost and capacity reasons) must be extremely conscious of the performance characteristics of PM and the challenges of using fast vs. slow memory in ways that best meet their performance demands. The key contribution of our research is a set of technologies that addresses these challenges in a bottom-up fashion. Since the real hardware is not yet available, we first implement a hardware emulator that can faithfully emulate the relative performance characteristics of DRAM and PM in a system with separate DRAM and emulated PM regions. We use this emulator to perform all of our evaluations. Next we explore system software support to enable low-overhead PM access by new and legacy applications. Towards this end, we implement PMFS, an optimized light-weight POSIX file system that exploits PM's byte-addressability to avoid overheads of block-oriented storage and enable direct PM access by applications (with memory-mapped I/O). To provide strong consistency guarantees, PMFS requires only a simple hardware primitive that provides software enforceable guarantees of durability and ordering of stores to PM. We demonstrate that PMFS achieves significant (up to an order of magnitude) gains over traditional file systems (such as ext4) on a RAMDISK-like PM block device. Finally, we address the problem of designing memory-based applications for systems with both DRAM and PM by extending our system software to manage both the tiers. We demonstrate for several representative large in-memory applications that it is possible to use a small amount of fast DRAM and large amounts of slower PM without a proportional impact to an application's performance, provided the placement of data structures is done in a careful fashion. To simplify the application programming, we implement a set of libraries and automatic tools (called X-Mem) that enables programmers to achieve optimal data placement with minimal effort on their part. Finally, we demonstrate the potentially large benefits of application-driven memory tiering with X-Mem across a range of applications.

No description available.

Web services have made many kinds of data and computing services available. However, to use web services often requires significant programming efforts and thus limits the people who can take advantage of them to only a small group of skilled programmers. In this dissertation, I will present a tool called Gneiss that extends the spreadsheet model to support four challenging aspects of using web services: programming two-way data communications with web services, creating interactive GUI applications that use web data sources, using hierarchical data, and using live streaming data. Gneiss contributes innovations in spreadsheet languages, spreadsheet user interfaces and interaction techniques to allow programming tasks that currently require writing complex, lengthy code to instead be done using familiar spreadsheet mechanisms. Spreadsheets are arguably the most successful and popular data tools among people of all programming levels. This work advances the use of spreadsheets to new domains and could benefit a wide range of users from professional programmers to end-user programmers.

Aujourd'hui, les plates-formes hétérogènes et partagées que sont les grilles de calcul sont omniprésentes. De plus, le besoin d'exécuter des applications parallèles complexes est croissant. Cette thèse vise à ordonnancer des applications représentées par des graphes de tâches modelables (dont le nombre de processeurs est fixé par l'ordonnanceur) sur des grilles de calcul en exploitant le maximum de parallélisme, utilisant efficacement les ressources, gérant l'hétérogénéité et le partage des ressources. Nous avons pour cela opté pour des heuristiques pragmatiques car, bien qu'elles n'offrent pas de garantie de performance, elles peuvent néanmoins conduire à de bonnes performances moyennes tout en construisant des ordonnancements en des temps relativement courts. La plupart des heuristiques existantes n'ordonnancent les applications parallèles mixtes qu'en milieu homogène et utilisent parfois inefficacement les ressources. Nous avons donc tout d'abord étudié différentes heuristiques dans le cas de plates-formes homogènes et proposé des améliorations visant à améliorer le compromis entre réduction du temps de complétion et efficacité. Nous avons ensuite introduit la gestion de l'hétérogénéité dans l'heuristique proposée et comparé ses performances à celles d'un algorithme garanti. Enfin, nous avons tenu compte du caractère partagé des grilles en gérant la concurrence entre applications. L'approche retenue consiste à limiter la quantité de ressources que chaque application peut utiliser pour construire son ordonnancement. Nous avons également proposé plusieurs stratégies de détermination de cette contrainte de ressources. / Today, computing grids, that are shared and heterogeneous platforms, are ubiquitous. Furthermore, the need to execute complex parallel applications is growing. The aim of this thesis is to schedule applications modeled by moldable task graphs (the number of processors allocated to each task is fixed by the scheduler) onto computing grids to exploit maximum parallelism, use efficiently the resources and manage the sharing of resources. We chose to design pragmatic heuristics because, although they do not guarantee performance, they can lead to good average performance while computing schedules in relatively short times. Almost all existing heuristics only schedule mixed parallel applications onto homogeneous platforms and they sometimes use inefficiently the resources. Thus, we first studied different scheduling heuristics in the case of homogeneous platforms and propose improvements to have a good compromise between the completion time and the efficiency. We then introduced management of heterogeneity in the proposed heuristic and compared its performances with those of a guaranteed algorithm. Finally, we have taken into account the fact that grids are shared by managing competition between applications. Our approach is to limit the amount of resources that each application can use to build its schedule. We also proposed different strategies to determine that resource constraint.

Les nitrures d'éléments II-IV ZnSiN2, ZnGeN2 et ZnSnN2 forment une famille de semi-conducteurs liés aux nitrures d'éléments III (le GaN et ses alliages contenant de l'aluminium ou de l'indium). Ils s'obtiennent par construction en remplaçant l'élément III (Ga) périodiquement par un élément II (Zn) puis par un élément IV (Si, Ge ou Sn), ses voisins de gauche et de droite dans le tableau périodique. La structure cristalline qui en résulte est très proche de celle du GaN wurtzite. Le ZnGeN2 présente un désaccord de maille avec le GaN inférieur à 1%. Sa largeur de bande interdite est de quelques pour cents identique à celle du GaN et le large décalage de bande entre le GaN et le ZnGeN2 permet la formation d'une hétérostructure de type II. Ces données ont ouvert la voie à l'étude théorique de l'intégration des matériaux II-IV-N2 dans les zones actives de LEDs GaN. Ces puits quantiques de type II pourraient contribuer à améliorer les propriétés d'émission à grandes longueurs d'onde (verte et au-delà) des émetteurs à base de GaN. L'alliage ZnSn{x}Ge{1-x}N2 (de x = 0 à x = 1) étant peu connu, l'objectif de la thèse est de réaliser une étude expérimentale du matériau sous forme de couches minces élaborées par pulvérisation cathodique magnétron réactive. Ses propriétés structurales, optiques et électriques sont étudiées au moyen de différentes méthodes d'analyse. Il paraît ainsi possible de moduler son paramètre de maille a (de 3.22 A à 3.41 A) ainsi que la largeur de la bande interdite (de 2.1 eV pour le ZnSnN2 à 3.0 eV pour le ZnGeN2) mais également ses propriétés électriques sur plusieurs ordres de grandeur. L'utilisation de substrats de GaN permet, en outre, une analyse de l'interface entre les deux matériaux et l'étude des effets de quasi-épitaxie. / The II-IV-nitrides ZnSiN2, ZnGeN2 and ZnSnN2 represent a semiconductors family close to the III-nitrides (GaN and its aluminum and indium containing alloys). They are obtained by replacing periodically the group III element (Ga) by a group II element (Zn) and by a group IV element (Si, Ge or Sn), its left and right neighbors in the periodic table. The crystalline structure of ZnGeN2 is therefore really close to the one of wurtzite GaN. They show a lattice mismatch smaller than 1 %. The band gap of ZnGeN2 is almost identical to GaN and their large band offset enables the design of a type II heterostructure. These data set the stage for the theoretical study of II-IV-N2 integration into the active zones of GaN LEDs. These type II quantum wells could contribute to enhance the emission properties of GaN-based light emitters at high wavelengths (green and beyond). The ZnSn{x}Ge{1-x}N2 alloy (with x = 0 to x = 1) being rather unknown, the objective of this thesis is the experimental study of sputtered thin films of this material. Its structural, optical and electrical properties are investigated through different analysis methods. It seems possible to adjust its lattice parameter a (from 3.22 A to 3.41 A) as well as its band gap (from 2.1 eV for ZnSnN2 to 3.0 eV for ZnGeN2) but also its electrical properties on several orders of magnitude. The use of GaN substrates enables the investigation of the interface between both materials and quasi-epitaxy effects.

Typescript (photocopy). / Digitized by Kansas Correctional Industries / Department: Electrical and Computer Engineering.

La réalité mixte apporte certains avantages aux applications industrielles. Elle peut, entre autres, faciliter la visualisation et validation de projets ou assister des opérateurs durant des tâches spécifiques. Les systèmes de projection (PBS), tels que les CAVE ou la réalité augmentée spatiale, fournissent un environnement de réalité mixte permettant une collaboration directe avec des utilisateurs externes. Dans cette thèse, nous visons à améliorer l'utilisation des systèmes de projection pour des applications industrielles en abordant deux défis majeurs: (1) améliorer les composantes techniques des PBS et (2) augmenter l'expérience utilisateur dans les PBS. En tant que premier défi technique, nous visons à améliorer les systèmes de suivi de mouvements optiques. Nous proposons une approche permettant d’élargir l’espace de travail de ces systèmes grâce à deux méthodes. La première permet de suivre les mouvements à partir d’une seule caméra tandis ce que la deuxième permet de contrôler les caméras et suivre les objets dans l’espace de travail. Le système qui en résulte fournit des performances acceptables pour des applications en réalité mixte tout en augmentant considérablement l'espace de travail. Un tel système de suivi de mouvement peut permettre de mieux exploiter le potentiel des systèmes de projection et d’élargir le champ possible des interactions. En tant que deuxième défi technique, nous concevons un casque « tout-en-un » pour la réalité augmentée spatiale mobile. Le casque rassemble à la fois un système de projection et un système de suivi de mouvements qui sont embarqués sur la tête de l'utilisateur. Avec un tel système, les utilisateurs sont capables de se déplacer autour d'objets tangibles et de les manipuler directement à la main tout en projetant du contenu virtuel par-dessus. Nous illustrons notre système avec deux cas d'utilisation industriels: le prototypage virtuel et la visualisation médicale. Enfin, nous abordons le défi qui vise à améliorer l’expérience utilisateur dans les PBS. Nous proposons une approche permettant d’incarner un personnage virtuel et d’augmenter la perception spatiale des utilisateurs dans les PBS. Pour ce faire, nous ajoutons l’ombre virtuelle des utilisateurs dans les systèmes de projection immersifs. L'ombre virtuelle est directement corrélée aux mouvements des utilisateurs afin qu'ils la perçoivent comme si c'était la leur. Nous avons effectué une expérience afin d'étudier l'influence de la présence de l'ombre virtuelle sur le comportement des utilisateurs. / Mixed reality brings some advantages to industrial applications. It can, among others, facilitate visualizing and validating projects or assist operators during specific tasks. Projection-based Systems (PBS), such as CAVEs or spatial augmented reality provide a mixed reality environment enabling straightforward collaboration with external users. In this thesis, we aim at improving the usage of PBS for industrial applications by considering two main challenges: (1) improving the technical components of PBS and (2) improving the user experience when using PBS. As a first technical challenge, we propose to address the improvement of the tracking component. We introduce an approach that enables increasing the workspace of optical tracking systems by using two methods. As a first method, we propose to use monocular tracking. As a second method, we propose to use controlled cameras that follow the targets across the workspace. The resulting system provides acceptable performances for mixed reality applications while considerably increasing the workspace. Such a tracking system can make it easier to use large projection-based displays and can widen the range of available interactions. As a second technical challenge, we design an “all-in-one” headset for mobile spatial augmented reality on tangible objects. The headset gathers both a projection and a tracking system that are embedded on the user’s head. With such a system, the users are able to move around tangible objects and to manipulate them directly by hand while projecting virtual content over them. We illustrate our system with two industrial use cases: virtual prototyping and medical visualization. Finally, we address the challenge that aims at improving the user experience when using PBS. We introduce a method that provides virtual embodiment and increases the spatial perception of the users when using PBS. To do so we add the user’s virtual shadow in immersive projection-based systems. The virtual shadow is dynamically mapped to users’ movements in order to make them perceive the shadow as if it was their own. We then carry out an experiment to study the influence of the presence of the virtual shadow on the user experience and behavior.

A thesis submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Doctor of Philosophy. Johannesburg, 2015.