No description available.

No description available.

Three different topics associated with their respective applications are proposed in this thesis. The first application is the implementation of a fast tagged sorter. A novel and high-speed realization of the tagged sorting algorithm is presented. Meanwhile, the problems to detect whether the queue is empty or full is also resolved without increasing any hardware cost. The second topic is focused on the implementation of a fast dynamic 64-bit comparator with small transistor count. The entire 64-bit comparator is composed of equality comparators and zero/one detectors, which are proposed by C.-F. Wu. The problem to handle a large fan-in requirement is also resolved in our design. The third topic is to carry out a power demand monitor system for factories. Not only can it monitor the factory¡¦s power network with a graphical user interface, but also can turn off the unessential equipments automatically when the total power consumed by the factory is larger than what was expected.

This work presents optimizations for modern capability machines and specifically for the CHERI architecture, a 64-bit MIPS instruction set extension for security, supporting fine-grained memory protection through hardware-enforced capabilities. The original CHERI model uses 256-bit capabilities to carry information required for various checks helping to enforce memory safety, leading to increased memory bandwidth requirements and cache pressure when using CHERI capabilities in place of conventional 64-bit pointers. In order to mitigate this cost, I present two new 128-bit CHERI capability formats, using different compression techniques, while preserving C-language compatibility lacking in previous pointer compression schemes. I explore the trade-offs introduced by these new formats over the 256-bit format. I produce an implementation in the L3 ISA modeling language, collaborate on the hardware implementation, and provide an evaluation of the mechanism. Another cost related to CHERI capabilities is the memory traffic increase due to capability-validity tags: to provide unforgeable capabilities, CHERI uses a tagged memory that preserves validity tags for every 256-bit memory word in a shadowspace inaccessible to software. The CHERI hardware implementation of this shadowspace uses a capability-validity-tag table in memory and caches it at the end of the cache hierarchy. To efficiently implement such a shadowspace and improve on CHERI’s current approach, I use sparse data structures in a hierarchical tag-cache that filters unnecessary memory accesses. I present an in-depth study of this technique through a Python implementation of the hierarchical tag-cache, and also provide a hardware implementation and evaluation. I find that validity-tag traffic is reduced for all applications and scales with tag use. For legacy applications that do not use tags, there is near zero overhead. Removing these costs through the use of the proposed optimizations makes the CHERI architecture more affordable and appealing for industrial adoption.

PSD-95 is an abundant scaffolding protein found in the postsynaptic densities (PSDs) of excitatory synapses throughout the mammalian brain, and plays a critical role in innate and learned behaviours. PSD-95 assembles with numerous other proteins, including glutamate receptors, adhesion molecules and signalling proteins, into postsynaptic supercomplexes that are then organised into nanoclusters that comprise the postsynaptic density of excitatory synapses. While the subcellular localisation of PSD-95 has been widely studied, much less is known about its turnover. In this thesis, I present novel insights into PSD-95 synthesis and degradation at synapses of cultured primary neurons gained using the HaloTag technology. The HaloTag consists of an engineered bacterial protein domain that covalently binds synthetic ligands labelled with fluorescent and affinity moieties. Hence, cells expressing proteins fused to the HaloTag can be used to study protein levels, complexes and turnover using these different ligands. This project was based upon a knock-in mouse line expressing the HaloTag fused to endogenous PSD-95 using gene targeting. After demonstrating that these mice were phenotypically normal and that PSD95-HaloTag fusion proteins normally assembled into supercomplexes in the PSD, hippocampal primary cultures were grown from this mouse line. Fluorescent HaloTag ligands were then used to label live neurons, allowing for the visualisation of PSD-95 at synapses by confocal microscopy. Next, I established a pulse-chase labelling method, where one ligand is used to label all existing PSD-95 first, before a second ligand can then be used to label any newly synthesised PSD-95. This allows for the identification and characterisation of subpopulations of PSD-95, which can be separately analysed. I find that PSD-95 has a half-life of 36 hours at synapses, consistent with previous literature. I was also able to observe synaptic heterogeneity in PSD-95 turnover, and classify synapses into types according to their PSD-95 expression profile. Finally, a range of chemical compounds known to modulate protein turnover and neuronal activity was applied over a 24-hour period, and their effects on PSD-95 turnover analysed. It was found that inhibiting either the proteasome or protein synthesis led to significant reductions in PSD-95 degradation as well as inhibiting PSD-95 synthesis. Thus, this project established a method offering a unique way of investigating the turnover of a specific, tagged protein, as well as gaining novel insights into the turnover of PSD-95 at individual synapses.

In recent years, increasing attentions are paid on developing exceptional technologies for efficiently processing massive collection of heterogeneous data generated by different kinds of sensors. While we have observed great successes of utilizing big data in many innovative applications, the need on integrating information poses new challenges caused by the heterogeneity of the data. In this thesis, we target at geo-tagged data, and propose a cloud based platform named City Digital Pulse (CDP), where a unified mechanism and extensible architecture are provided to facilitate the various aspects in big data analysis, ranging from data acquisition to data visualization. We instantiate the proposed system using multi-model data collected from two social platforms, Twitter and Instagram, which include plenty of geo-tagged messages. Data analysis is performed to detect human affections from the user uploaded content. The emotional information in big social data can be uncovered by using a multi-dimension visualization interface, based on which users can easily grasp the evolving of human affective status within a given geographical area, and interact with the system. This offers costless opportunities to improve the decision making in many critical areas. Both the proposed architecture and algorithm are empirically demonstrated to be able to achieve real-time big data analysis.

Only two percent of new books released in Germany are professionally edited for visually impaired people. However, more and more print publications are made available to the public in digital formats through online content delivery platforms like “libreka!”. The automatic conversion of such contents into DAISY would considerably increase the number of publications available in accessible formats. Still, most data available on “libreka!” is published as non-tagged PDF. In this paper, we examine the potential for automatic conversion of “libreka!”-based content into DAISY, while also analyzing the potentials and limitations of current conversion tools.

Chez les bactéries en contact direct avec leur milieu, la transcription des gènes et la synthèse des protéines sont régulées de manière efficace à chaque changement des paramètres environnementaux afin de permettre la survie cellulaire. Dans le cas des bactéries commensales de l’intestin, ces régulations doivent aussi permettre les interactions symbiotiques et la colonisation dont les mécanismes moléculaires, encore peu connus, sont probablement liés, entre autres, à la surface des bactéries (molécules exposées et sécrétées…). Lactobacillus casei, bactérie commensale, possède environ 330 gènes prédits comme intervenant dans la composition et la fonctionnalité de la surface cellulaire. Afin d’avoir une vue globale de la totalité des gènes qui interviennent dans l’établissement de L. casei dans l’intestin, une approche de génétique inverse a été réalisée. Pour cela, une banque de mutants aléatoires étiquetés de L. casei par « Signature-Tagged Mutagenesis » a été créée puis annotée et réorganisée grâce au séquençage des régions d’insertion du transposon. Les mutants ont été criblés quant à leur capacité à s’établir dans l’anse iléale ligaturée de lapin et quantifiés par qPCR. Parmi les 47 gènes identifiés comme étant impliqués dans l’établissement in vivo, trois gènes en opéron codant pour un système à deux composants et une « penicillin-binding protein » ont été caractérisés. Ces trois gènes sont impliqués dans la modulation de la surface cellulaire et plus particulièrement dans la régulation des hydrolases du peptidoglycane qui sont nécessaires à la protection de la bactérie dans l’environnement intestinal. / In bacteria which are in direct contact with their environment, genes transcription and proteins synthesis are efficiently regulated at each change of environmental parameters to allow cell survival. For intestinal commensal bacteria, these regulations must also allow symbiotic interactions and colonization whose molecular mechanisms, so far little known, are probably related, among others, to the bacteria surface (molecules exposed and secreted…). Lactobacillus casei, a commensal bacterium, has about 330 predicted genes involved in the composition and functionality of the cell surface. To have a global view of the whole genes involved in the establishment of L. casei in the gut, a reverse genetics approach was performed. For that, a library of L. casei random labeled-mutants by Signature-Tagged Mutagenesis was generated then annotated and reassembled thanks to the sequencing of transposon insertion sites. Mutants were screened for their ability to establish themselves in the rabbit ligated ileal loop and quantified by qPCR. Among the 47 genes identified as involved in the in vivo establishment, three genes in an operon encoding a two-component system and a penicillin-binding protein were characterized. These three genes are involved in the cell surface modulation and particularly in the regulation of peptidoglycan hydrolases which are required for the bacteria protection in the intestinal environment.

Only two percent of new books released in Germany are professionally edited for visually impaired people. However, more and more print publications are made available to the public in digital formats through online content delivery platforms like “libreka!”. The automatic conversion of such contents into DAISY would considerably increase the number of publications available in accessible formats. Still, most data available on “libreka!” is published as non-tagged PDF. In this paper, we examine the potential for automatic conversion of “libreka!”-based content into DAISY, while also analyzing the potentials and limitations of current conversion tools.

International Telemetering Conference Proceedings / October 17-20, 1988 / Riviera Hotel, Las Vegas, Nevada / The growing complexity of missle, aircraft, and space vehicle systems, along with the advent of fly-by-wire and ultra-high performance unstable airframe technology has created an exploding demand for real time processing power. Recent VLSI developements have allowed addressing these needs in the design of a multi-processor subsystem supplying 10 MIPS and 5 MFLOPS per processor. To provide up to 70 MIPS a Digital Signal Processing subsystem may be configured with up to 7 Processors. Multiple subsystems may be employed in a data processing system to give the user virtually unlimited processing power. Within the DSP module, communication between cards is over a high speed, arbitrated Private Data bus. This prevents the saturation of the system bus with intermediate results, and allows a multiple processor configuration to make full use of each processor. Design goals for a single processor included executing number system conversions, data compression algorithms and 1st order polynomials in under 2 microseconds, and 5th order polynomials in under 4 microseconds. The processor design meets or exceeds all of these goals. Recently upgraded VLSI is available, and makes possible a performance enhancement to 11 MIPS and 9 MFLOPS per processor with reduced power consumption. Design tradeoffs and example applications are presented.