The aim of this thesis was to study and subsequently process a comparison of older and newer SIMD processing units of modern microprocessors on the x86-64 platform. The thesis provides an overview of the fastest computations of vector operations with matrices and vectors, including corresponding source codes. Furthermore, the thesis is focused on authenticated encryption, specifically on block cipher AES operating in Galois Counter Mode, and on a discussion of possibilities of instruction sets for cryptographic support.

Vi har i detta projekt arbetat med att utforma ett robust och säkert system som är byggt med hjälp av klient-server arkitekturen åt Tweakers HB. Tweakers HB är ett nytt företag som utvecklar allt från mobila till stationära applikationer. Produkten som utvecklats är ett ordbaserat frågesportsspel som har multiplayer funktionalitet. Applikationen/klienten var utvecklat för mobiler som använder Android plattformen och server bestod av ett REST API och en MYSQL databas. Klient-sidan byggdes upp med en anpassad tillämpning av MVC mönstret och använde SQLite för att spara data lokalt. Målet med projektet kommer vara att utvecklat ett system som kommer att ha en lång livslängd som även kan byggas ut i framtiden. För att kunna uppnå målet med produkten så användes utvecklingsmetoden XP(Extreme Programming) och test-driven utveckling. Företaget la vissa krav på tekniker som skulle användas men i stort sett så låg ansvaret på utvecklingsgruppens att hitta tekniker och biliotek för att uppnå kraven. En stor del av kraven blev uppfyllda. / We have in this project worked on designing a robust and reliable system that is built using the client-server architecture for Tweakers HB. Tweakers HB is a new company that develops everything from mobile to desktop applications. The product developed is a word based quiz game with multiplayer functionality. The application/client was developed for mobile phones using the Android platform and the server consisted of a REST API and a MySQL database. The client-side was built using a custom implementation of the MVC pattern and used SQLite to store data locally. The goal of the project was to develop a system that will have a long life span that could also be expanded in the future. In order to achieve the goal of the product XP (Extreme Programming) development methodology and test-driven development was used. The company put certain requirements for technologies that could be used but basically the responsibility fell upon development team to find techniques and libraries to achieve the requirements. Much of demands where satisfied.

Målet med detta projekt var att utforma en kommunikation och databassystem för ett multiplayer ordbaserat spel åt Tweakers HB. Tweakers HB är ett produktbolag som utvecklar mobila applikationer där Wordhunch är en utav deras produkter. Systemet är en webbaserad API som stöds av en databas system för att kunna hantera olika aspekter av ett multiplayer ordbaserat spel. API: et kommer att användas av en Android applikation som utvecklas av andra utvecklare på Tweakers HB. Kommunikationssystemet har tagits fram med Extreme Programming samt testdriven utveckling. Utvecklingen skedde med välkända och beprövade metoder för att kunna säkerställa att ett system med hög kvalitet levereras. Båda system utnyttjar gamla och vältestade verktyg så som MySQL för databashantering medan REST-arkitektur används för kommunikationssystem. I slutet av utvecklingen uppfylldes alla ställda krav. System klarar av prestandakravet enligt de tester som utfördes. Under testning har flera komplikationer uppstått som inte var kända under utvecklingen eller före som t.ex. ta fram det korrekta ord när ett felstavat ord har skickats in. / The goal of the project was to design a communication and database system for a multiplayer word based game for Tweakers HB. Tweakers HB is a company that develops mobile applications, where WordHunch is one of their products. The product itself is a web based API which is supported by a database system for handling different areas of a multiplayer based mobile game. The API will be used by the other developers of Tweakers HB for communication between Server and multiple Android and iOS mobiles. The system is a fully accomplished system. The system has been produced using Extreme programming. Development was done with well-known and proven methods to ensure that a system of high-quality was delivered. Both systems utilizes old and well tested products like MySQL for database management while REST architecture style for API system. At the end of the project all requirements were met. The system passed all the required tests conducted and will be able to handle large quantity of incoming traffic from Android and iOS mobile devices, according to tests. During testing multiple issues came up which were neither known before the production or under production for ex. retrieving the correct word when a misspelled word has been submitted.

Les modèles climatiques ont beaucoup progressé pour représenter les nuages. Pourtant la réponse et la rétroaction nuageuse demeure très différente d'un modèle à l'autre, et reste la principale source d'incertitude pour la sensibilité climatique prédite par les modèles de climat, et limite la fiabilité sur les projections du changement climatique dû au forçage anthropique. Il est donc crucial d'améliorer la représentation des processus nuageux dans les modèles climatiques. L'évaluation des nuages nécessite des observations précises. Jusqu'à récemment, des observations de plusieurs aspects fondamentaux des nuages comme la distribution tridimensionnelle des nuages existaient uniquement très grossièrement et obtenus de manière très indirecte par les satellites de télédétection passive (i.e. CERES, ERBE, ScaRab, ISCCP) qui mesurent les flux radiatifs au sommet de l'atmosphère. Les observations A-Train constituent des outils exceptionnels pour caractériser les propriétés nuageuses. L'objet de cette thèse est de tirer profit des observations de l'A-train afin d'évaluer la description des nuages simulée par les modèles climatiques. Nous utilisons le radiomètre CERES pour estimer l'effet radiatif des nuages, les radiomètres PARASOL et MODIS qui mesurent la réflectance, analysée ici comme un proxy de l'épaisseur optique des nuages et le lidar CALIPSO qui fournit des informations précises sur la distribution verticale des nuages. Les données co-localisées et analysées statistiquement constituent une occasion exceptionnelle de contraindre simultanément les propriétés radiatives des nuages et leur distribution tridimensionnelle. Le modèle du climat évalué est le LMDZ ainsi qu'une nouvelle version du modèle en cours de développement, où des nouvelles paramétrisations du bloc couche-limite/convection/nuages est testée. La méthode de comparaison des sorties des modèles climatiques aux grandeurs observées utilise d'une part le simulateur COSP (CFMIP Observation Simulator Package) qui comprend SCOPS, le simulateur lidar et le simulateur PARASOL et d'autre part les jeux des données (CFMIP-OBS) construits pour être compatibles avec les simulateurs. Nous étudions les propriétés nuageuses dans les tropiques par régime de circulation, et en classant les nuages par régions. Une nouvelle méthode a été développée : les observations sont analysées à haute résolution (spatiale et temporelle) au lieu des moyennes mensuelles et saisonnières utilisées habituellement afin de se placer à une échelle aussi proche que possible de celle des processus nuageux. Cette analyse a permis de contraindre les paramétrisations développées pour représenter les nuages et révéler des biais dans les deux versions du LMDZ. Des compensations d'erreurs ont été identifiées (i) sur la distribution verticale des nuages : la couverture nuageuse des nuages hauts et surestimée alors que les nuages bas et moyens sont significativement sous-estimés, (ii) entre la couverture nuageuse et l'épaisseur optique : la couverture nuageuse totale est sous-estimée mais les nuages qui se forment ont une épaisseur optique très élevée ce qui aboutit à une simulation correcte des flux au sommet de l'atmosphère par le modèle.

No description available.

The Galois/Counter Mode of Operation (GCM), recently standardized by NIST, simultaneously authenticates and encrypts data at speeds not previously possible for both software and hardware implementations. In GCM, data integrity is achieved by chaining Galois field multiplication operations while a symmetric key block cipher such as the Advanced Encryption Standard (AES), is used to meet goals of confidentiality. Area optimization in a number of proposed high throughput GCM designs have been approached through implementing efficient composite Sboxes for AES. Not as much work has been done in reducing area requirements of the Galois multiplication operation in the GCM which consists of up to 30% of the overall area using a bruteforce approach. Current pipelined implementations of GCM also have large key change latencies which potentially reduce the average throughput expected under traditional internet traffic conditions. This thesis aims to address these issues by presenting area efficient parallel multiplier designs for the GCM and provide an approach for achieving low latency key changes. The widely known Karatsuba parallel multiplier (KA) and the recently proposed Fan-Hasan multiplier (FH) were designed for the GCM and implemented on ASIC and FPGA architectures. This is the first time these multipliers have been compared with a practical implementation, and the FH multiplier showed note worthy improvements over the KA multiplier in terms of delay with similar area requirements. Using the composite Sbox, ASIC designs of GCM implemented with subquadratic multipliers are shown to have an area savings of up to 18%, without affecting the throughput, against designs using the brute force Mastrovito multiplier. For low delay LUT Sbox designs in GCM, although the subquadratic multipliers are a part of the critical path, implementations with the FH multiplier showed the highest efficiency in terms of area resources and throughput over all other designs. FPGA results similarly showed a significant reduction in the number of slices using subquadratic multipliers, and the highest throughput to date for FPGA implementations of GCM was also achieved. The proposed reduced latency key change design, which supports all key types of AES, showed a 20% improvement in average throughput over other GCM designs that do not use the same techniques. The GCM implementations provided in this thesis provide some of the most area efficient, yet high throughput designs to date.

The Galois/Counter Mode of Operation (GCM), recently standardized by NIST, simultaneously authenticates and encrypts data at speeds not previously possible for both software and hardware implementations. In GCM, data integrity is achieved by chaining Galois field multiplication operations while a symmetric key block cipher such as the Advanced Encryption Standard (AES), is used to meet goals of confidentiality. Area optimization in a number of proposed high throughput GCM designs have been approached through implementing efficient composite Sboxes for AES. Not as much work has been done in reducing area requirements of the Galois multiplication operation in the GCM which consists of up to 30% of the overall area using a bruteforce approach. Current pipelined implementations of GCM also have large key change latencies which potentially reduce the average throughput expected under traditional internet traffic conditions. This thesis aims to address these issues by presenting area efficient parallel multiplier designs for the GCM and provide an approach for achieving low latency key changes. The widely known Karatsuba parallel multiplier (KA) and the recently proposed Fan-Hasan multiplier (FH) were designed for the GCM and implemented on ASIC and FPGA architectures. This is the first time these multipliers have been compared with a practical implementation, and the FH multiplier showed note worthy improvements over the KA multiplier in terms of delay with similar area requirements. Using the composite Sbox, ASIC designs of GCM implemented with subquadratic multipliers are shown to have an area savings of up to 18%, without affecting the throughput, against designs using the brute force Mastrovito multiplier. For low delay LUT Sbox designs in GCM, although the subquadratic multipliers are a part of the critical path, implementations with the FH multiplier showed the highest efficiency in terms of area resources and throughput over all other designs. FPGA results similarly showed a significant reduction in the number of slices using subquadratic multipliers, and the highest throughput to date for FPGA implementations of GCM was also achieved. The proposed reduced latency key change design, which supports all key types of AES, showed a 20% improvement in average throughput over other GCM designs that do not use the same techniques. The GCM implementations provided in this thesis provide some of the most area efficient, yet high throughput designs to date.

There are many unsolved questions about the future of climate, and most of them are due to lack of knowledgeabout the complex system of atmosphere, but still there are models that produce relatively realistic projectionsof the future although there are uncertainties in the presentation of them, and that's where statistical methodscould be of help. Here a short introduction is given to the projection of future climate with GCM ensembles andthe uncertainties about them, the emerging probabilistic approach, as well as the REA (Reliability EnsembleAverage) method for measuring the reliability of the model projections. In order to have an impression of theresults of the GCM ensemble results and their uncertainties the results of the weather forecast over a time periodof one year in three dierent cities of Sweden is studied as well.

The increasing global air temperature will trigger changes in the global mean water vapor, precipitation patterns and evapotranspiration, which further leads to changes, for instance, instream flow, groundwater flow and soil moisture. Projections of future changes in thehydrological regime of the Aral Sea Drainage Basin (ASDB) in Central Asia are however highlyuncertain, due to complexities of natural and engineered water systems of the basin. The AmuDarya River Delta (ADRD) is vital to the water budget of the Large Aral Sea, the livelihood inUzbekistan and Turkmenistan, as well as the surrounding riparian ecosystem. This study attemptsto investigate responses of river flow in the Aral Sea Drainage Basin and key riparian vegetationspecies (of the so-called Tugai community) in the Amu Darya River Delta to projected futureclimate change. Results from hydrological model and outputs from multi-GCM predictions providea basis for conducting more robust quantitative analysis of possible future hydro-climatic changesin the Amu Darya River Basin. A qualitative synthesis of the suitability of Tugai is furthermoreperformed in order to increase the knowledge of the riparian vegetation status under thechanging hydro-climatic conditions. The results show that the averaged temperature in the ASDBis likely to continuously increase and yield a total increase of about 2 °C ~ 5°C by 2100. Thechange trend of the annual regional precipitation of 2100 is relatively unclear, with estimatesranging from 50 mm lower than today to 75 mm higher than today. Modeled ensemble means (EM)river flow, obtained from hydrological modeling of climate output from multi-GCM projections,converge on showing future decreases in river runoff (R). Projected absolute R may decrease tozero around 2100, implying no surface flow and a dry out near the river outlet. The relationship ofwater flux between upstream and downstream will be changed dramatically due to climatechange. More specifically, R of the upstream region will decrease, and it is likely to becomeinsufficient for feeding downstream river reaches as it used to. The decreased river flow in thedelta may accelerate the desertification and salinization processes. Consequently, speciestransitions may occur, along with degradations of the existing Tugai communities. Theuncertainties of hydro-climatic change projections to some extent hinder the understanding of thedynamic hydrological-climatic-ecological system. However, the detailed responses of the delta toclimate change based on multiple qualitative and quantitative analyses provide an important basisfor the formulation of more robust forecasts on the future ecological development in the ADRD, and further for recommendations of measures to mitigate the ecosystem’s deterioration under achanging climate.

The influence of anthropogenic emissions on aerosol distributions and the hydrological cycle are examined with a focus on monsoon precipitation over the Indian subcontinent, during January 2001 to December 2005, using the European Centre for Medium-Range Weather Forecasts-Hamburg (ECHAM5.5) general circulation model extended by the Hamburg Aerosol Module (HAM). The seasonal variability of aerosol optical depth (AOD) retrieved from the MODerate Resolution Imaging Spectroradiometer (MODIS) on board the Terra and Aqua satellite is broadly well simulated (R 0.6–0.85) by the model. The spatial distribution and seasonal cycle of the precipitation observed over the Indian region are reasonably well simulated (R 0.5 to 0.8) by the model, while in terms of absolute magnitude, the model underestimates precipitation, in particular in the south-west (SW) monsoon season. The model simulates significant anthropogenic aerosol-induced changes in clear-sky net surface solar radiation (dimming greater than -7 W m-2), which agrees well with the observed trends over the Indian region. A statistically significant decreasing precipitation trend is simulated only for the SWmonsoon season over the central-north Indian region, which is consistent with the observed seasonal trend over the Indian region. In the model, this decrease results from a reduction in convective precipitation, where there is an increase in stratiform cloud droplet number concentration (CDNC) and solar dimming that resulted from increased stability and reduced evaporation. Similarities in spatial patterns suggest that surface cooling, mainly by the aerosol indirect effect, is responsible for this reduction in convective activity. When changes in large-scale dynamics are allowed by slightly disturbing the initial state of the atmosphere, aerosol absorption in addition leads to a further stabilization of the lower troposphere, further reducing convective precipitation.