Nach einer Einfürung in SPEC und deren Berwertungs- verfahren wird die Art und Weise der SPEC-Leistungsmessungen eingehender beleuchtet. Der Hauptteil beinhaltet Durchführung sowie Auswertung von SPEC-Benchmarks an 5 Workstations. Die dabei gewonnenen Ergebnisse werden mit den offiziell verbreiteten SPEC-Werten verglichen und diskutiert.

A collection of benchmark examples is presented for the numerical solution of continuous-time algebraic Riccati equations. This collection may serve for testing purposes in the construction of new numerical methods, but may also be used as a reference set for the comparison of methods.

This is the second part of a collection of benchmark examples for the numerical solution of algebraic Riccati equations. After presenting examples for the continuous-time case in Part I, our concern in this paper is discrete-time algebraic Riccati equations. This collection may serve for testing purposes in the construction of new numerical methods, but may also be used as a reference set for the comparison of methods.

Infiniband wird zunehmend als Verbindungsnetzwerk für Cluster eingesetzt. Dadurch wird es nötig existierende Bibliotheken für parallele Programmiersprachen an das neue Netzwerk bestmöglich anzupassen. Ein wichtiger Bestandteil paralleler Programmiersprachen sind dabei kollektive Operationen, die es erfordern, eine Nachricht von einem Knoten zu vielen anderen oder auch von vielen Knoten an einen einzelnen zu senden. Um herauszufinden, welche Verbindungsarten und Operationen am besten für diese kollektiven Operationen geeignet sind, wurde ein Benchmark entwickelt. Ziel dieser Studienarbeit ist es, dieses Programm zu erweitern, auf einem Cluster zu testen und die Ergebnisse auszuwerten.

The recent growth in sensor technology allows easier information gathering in real-time as sensors have grown smaller, more accurate, and less expensive. The resulting data is often in a geo-stream format continuously changing input with a spatial extent. Researchers developing geo-streaming management systems (GSMS) require a benchmark system for evaluation, which is currently lacking. This thesis presents GSMark, a benchmark for evaluating GSMSs. GSMark provides a data generator that creates a combination of synthetic and real geo-streaming data, a workload simulator to present the data to the GSMS as a data stream, and a set of benchmark queries that evaluate typical GSMS functionality and query performance. In particular, GSMark generates both moving points and evolving spatial regions, two fundamental data types for a broad range of geo-stream applications, and the geo-streaming queries on this data.

This study proposes a performance comparison between the different structures of NoSQL databases; document, key-value, column and graph. A second study is also conducted, when looking at performance comparison between three different NoSQL databases, all of the same structure; document based, the databases that are tested here are; MongoDB, OrientDB and Couchbase. Performance tests are conducted using a benchmarking tool YCSB (Yahoo! Cloud Serving Benchmark), and by looking at time to execute and throughput (operations/ second). Beside benchmarking literature reviews are conducted to be able to understand the different NoSQL structures, and to elaborate our benchmarking results. Every NoSQL structure and database in our benchmark is tested in the same way, a loading phase of 1k, 10k and 100k entries, and a running phase with a workload of approximately 50% reads and 50% updates with 1k, 10k and 100k operations. The finding of this study is that there are differences in performance, both between different structures and between same structured NoSQL databases. Document based OrientDB was the highest performing database at high volumes of data, and key-value store database Redis performed best at low volumes of data. Reasons for performance differences are both linked to specific trademarks of the structural orientation, the usage of the specific attributes of CAP theorem, storage type and development language.

The purpose of this thesis is to describe the relationship between surface water quality and land use. Water management recommendations will be divulged based upon the interaction of lake water quality and land use. The methodology developed for this research applied Data Envelopment Analysis (DEA), a performance measurement tool, to evaluate lake water quality in relation to surrounding land use. Lake performance ratings were generated by DEA software that assessed multiple variables describing surface water nutrient loads and surrounding land use. Results from this analysis revealed a significant trend between lake water quality and land use within the study area. Lakes located within a two mile radius of more naturally preserved land area typically attained higher performance ratings than lakes located within a two mile radius of less naturally preserved land area. The spatial quantity of naturally preserved land influenced lake nutrient concentrations. Also, lake performance ratings generally declined in two mile radius delineations that contained less naturally preserved land area indicating a direct relationship between natural land area and lake performance.

La mesure de performance est utilisée dans tous les systèmes informatiques pour garantir la meilleure performance pour le plus faible coût possible. L'établissement d'outils de mesures et de métriques a permis d'établir des bases de comparaison entre ordinateurs. Bien que le monde de la carte à puce ne fasse pas exception, les questions de sécurité occupent le devant de la scène pour celles-ci. Les efforts allant vers une plus grande ouverture des tests et de la mesure de performance restent discrets. Les travaux présentés ici ont pour objectif de proposer une méthode de mesure de la performance dans les plates-formes Java Card qui occupent une part considérable du marché de la carte à puce dans le monde d’aujourd’hui. Nous étudions en détails les efforts fournis par d'autres auteurs sur le sujet de la mesure de performance et en particulier la mesure de performance sur les cartes à puce. Un grand nombre de ces travaux restent embryonnaires ou ignorent certains aspects des mesures. Un des principaux défauts de ces travaux est le manque de rapport entre les mesures effectuées et les applications généralement utilisées dans les cartes à puce. Les cartes à puce ont par ailleurs des besoins importants en termes de sécurité. Ces besoins rendent les cartes difficiles à analyser. L'approche logique consiste à considérer les cartes à puce comme des boites noires. Après l'introduction de méthodologies de mesures de performance pour les cartes à puce, nous choisirons les outils et les caractéristiques des tests que nous voulons faire subir aux cartes, et nous analyserons la confiance à accorder aux données ainsi récoltées. Enfin une application originale des cartes à puce est proposée et permet de valider certains résultats obtenus. / Performance measurements are used in computer systems to guaranty the best performance at the lowest cost. Establishing measurement tools and metrics has helped build comparison scales between computers. Smart cards are no exception. But the centred stage of the smart card industry is mostly busy with security issues. Efforts towards a better integration of performance tests are still modest. Our work focused on a better approach in estimating the execution time within Java Card platforms. Those platforms constitute a big part of the modern smart card market share especially with regards to multi-applicative environments. After introducing some methodologies to better measure the performance of Java Cards, we detail the tools and the tests that we mean to use on smart cards. We will thereafter analyze the data obtained in this way. Finally, an original application for smart cards is proposed. We used it to validate some points about the results.

Bakgrund . Moderna spel blir mer krävande på hårdvaran och för att motverka detta utvecklas nya tekniker för att lätta på dessa krav. En sådan optimeringsteknik är Variable Rate Shading (VRS), som ingår i DirectX 12 API. Det tillåter utvecklare att variera kvaliteten på delar av ramen för att förbättra prestanda. Hur effektiv VRS är, verkar variera eftersom olika benchmark-testare får olika resultat. Detta beror med största sannolikhet på de olika scenmiljöer som används i testerna. Mål . För att ytterligare utöka de miljöer som används VRS bench-tester kommer denna studie att fokusera på att mäta och utvärdera hos VRS i en lätt miljö som skiljer sig från andra. Metoder . Metoden består av att utveckla en lättviktig Direct3D 12-applikation, implementera VRS-tekniken och mäta prestanda. För en tydlig utvärdering genomförs flera tester som mäter bildrutetid, bildhastighet och ritar samtalshastighet vid de olika inställningarna med hjälp av VRS-tekniken vid olika upplösningar över 1000 iterationer. Resultat . Genom att mäta bildrutetid, bildfrekvens och dragsamtalshastighet med VRS var det möjligt att samla in prestationsdata som visa i denna studie. Studien visar den genomsnittliga prestandan med 1x1, 2x2 och 4x4 skugghastigheter vid 480p, 1080p och 2160p upplösning. Medeldata jämfördes mellan skuggningshastigheter och upplösningar för att undersöka korrelationen och avvikelsen. Som förväntat visas generella prestandaförbättringar vid användning av VRS. Vissa inställningar visade dock inkonsekvens i avvikelser mellan skuggningshastigheter, och andra visade försämrad prestanda. Slutsatser . Slutsatsen från denna studie tider på att VRS förbättrar prestanda även i lätta applikationer, inom rimliga gränser. Prestandavinsten var dock av lägre grad vid jämförelse med andra benchmarktester. Detta tyder på att VRS är mer användbar i mer krävande miljöer.

A state machine model can turn a complex behavioural system into a more accessible graphical model, and can improve the way people work with system design by making it easier to communicate and understand the system. The clear structure of a state machine model enables automatic generation of well structured, and consequently readable, and maintainable code. There are many known implementations and even plenty of commercial software available for working with state machines, and the goal of this project is to compare and discuss some of these implementations in the context of Ericsson's demands for run time, memory usage, scalability, readability, and maintainability. More specifically, the project focuses on the state machine models specified in the UML (the unified modeling language [15]), utilizing UML's associated markup language, XMI, to go from graphical model to generated C code. The resulting C code is primarily a code skeleton which only provides the basic behaviour of transitioning between states given a specific event, it is expected that the developers manually implement additional features themselves. The examined implementations are: Nested Switch, Array of Structs, Function Pointers, Basic State Pattern, State-Table Pattern, and Hierarchical State Pattern. Additionally, the project investigates how multiple state machines can communicate and work together as interacting state machines. And finally, to showcase how a state machine implementation can be maintainable, we develop an iterative code editor that can edit already operating and manually modified state machine implementations. The implementations are tested on a case study example provided by Ericsson, aimed to represent a sort of typical state machine design when it comes to number of states and events. The implementations are further tested with randomly generated state machines, to examine their scalability properties. In our opinion the results favour the Array of Structs and Basic State Pattern implementations and the choice depends on the optimisation used and the priority between run time and memory.