Notre objectif est la conception d'algorithmes d'ordonnancement temps réel en-ligne faisables (i.e., garantissant le respect des contraintes temporelles imposées au systèmes) optimisant 1) l'utilisation de la plate-forme d'exécution (i.e., utiliser au mieux le potentiel de la plate-forme d'exécution tout en garantissant le respect des contraintes temporelles imposées au système) et/ou 2) optimisant des critères de performances<br />propres à l'application.<br /> <br />Deux cas ont été analysés : le cas de tâches indépendantes périodiques s'exécutant sur un processeur et le cas de flux de messages indépendants périodiques sur un réseau de terrain avec accès au médium priorisé.<br /><br />Nous avons donc proposé pour traiter les deux problèmatiques abordées :<br />1) des méthodes de configurations permettant d'optimiser l'utilisation de la plate-forme d'exécution en fixant d'une manière appropriée les paramètres : des politiques ou des activités du système considéré. Deux études ont été conduites dans ce cadre~: l'allocation des ``offsets'' dans les systèmes ``offset free'' et l'allocation de priorités, de politiques et de quantum dans<br />les systèmes conformes au standard Posix~1003.1b,<br />2) une nouvelle classe de politiques d'ordonnancement permettant<br />d'optimiser des critères de performances propres à l'application.

Modern technologies are growing and Voice over Internet Protocol (VoIP) technology is able to function in heterogeneous networks. VoIP gained wide popularity because it offers cheap calling rates compared to traditional telephone system and the number of VoIP subscribers has increased significantly in recent years. End users need reliable and acceptable call quality in real time communication with best Quality of Service (QoS). Server complexity is increasing to handle all client requests simultaneously and needs huge processing power. VoIP Servers will increase processing power but the engineering tradeoff needs to be considered e.g. increasing hardware will increase hardware complexity, energy consumption, network management, space requirement and overall system complexity. Modern System-on-Chip (SoC) uses multiple core technology to resolve the complexity of hardware computation. With enterprises needing to reduce overall costs while simultaneously improving call setup time, the amalgamation of VoIP with SoC can play a major role in the business market. The proposed VoIP Server model with multiple processing capabilities embedded in it is tailored for multicore hardware to achieve the required result. The model uses SystemC-2.2.0 and TLM-2.0 as a platform and consists of three main modules. TLM is built on top of SystemC in an overlay architectural fashion. SystemC provides a bridge between software and hardware co-design and increases HW &amp; SW productivity, driven by fast concurrent programming in real time. The proposed multicore VoIP Server model implements a round robin algorithm to distribute transactions between cores and clients via Load Balancer. Primary focus of the multicore model is the processing of call setup time delays on a VoIP Server. Experiments were performed using OpenSIP Server to measure Session Initiation Protocol (SIP) messages and call setup time processing delays. Simulations were performed at the KTH Ferlin system and based on the theoretical measurements from the OpenSIP Server experiments. Results of the proposed multicore VoIP Server model shows improvement in the processing of call setup time delays.

Edge computing, as an approach to leveraging computation capabilities located in different places, is widely deployed in the industry nowadays. With the development of edge computing, many big companies move from the traditional monolithic software architecture to the microservice design. To provide better performance of the applications which contain numerous loosely coupled modules that are deployed among multiple clusters, service routing among multiple clusters needs to be effective. However, most existing solutions are dedicated to static service routing and load balancing strategy, and thus the performance of the application cannot be effectively optimized when network condition changes.To address the problem mentioned above, we proposed a dynamic weighted round robin algorithm and implemented it on top of the cutting edge service mesh Istio. The solution is implemented as a Docker image called RoutingAgent, which is simple to deployed and managed. With the RoutingAgent running in the system, the weights of the target routing clusters will be dynamically changed based on the detected inter-cluster network latency. Consequently, the client-side request turnaround time will be decreased.The solution is evaluated in an emulated environment. Compared to the Istio without RoutingAgent, the experiment results show that the client-side latency can be effectively minimized by the proposed solution in the multicluster environment with dynamic network conditions. In addition to minimizing response time, emulation results demonstrate that loads of each cluster are well balanced. / Edge computing, som ett tillvägagångssätt för att utnyttja beräkningsfunktioner som finns på olika ställen, används i stor utsträckning i branschen nuförtiden. Med utvecklingen av kantdatabasen flyttar många stora företag från den traditionella monolitiska mjukvaruarkitekturen till mikroserviceteknik. För att ge bättre prestanda för de applikationer som innehåller många löst kopplade moduler som distribueras bland flera kluster, måste service routing bland flera kluster vara effektiva. De flesta befintliga lösningarna är dock dedikerade till statisk service-routing och belastningsbalanseringsstrategi, vilket gör att programmets prestanda inte effektivt kan optimeras när nätverksförhållandena ändras.För att ta itu med problemet som nämnts ovan föreslog vi en dynamisk viktad round robin-algoritm och implementerade den ovanpå den avancerade servicenätverket Istio. Lösningen implementeras som en Docker-bild som heter RoutingAgent, som är enkel att distribuera och hantera. Med agenten som körs i systemet ändras vikten av målruteringsklustret dynamiskt baserat på den upptäckta interklusternätets latens. Följaktligen kommer klientsidans begäran om omställningstid att minskas.Lösningen utvärderas i en emulerad miljö. Jämfört med Istio utan agent visar experimentresultaten att klientens latentitet effektivt kan minimeras av den föreslagna lösningen i multicluster-miljö med dynamiska nätverksförhållanden. Förutom att minimera responstid visar emuleringsresultat att belastningar i varje kluster är välbalanserade.

The growing interest towards wireless communication services over the recent years has increased the demand for radio spectrum. Inefficient spectrum management together with the scarcity of the radio spectrum is a limiting factor for the development of modern wireless networks. As a solution, the idea of cognitive radio networks (CRNs) is introduced to use licensed spectrum for the benefit of the unlicensed secondary users. However, the preemptive priority of the licensed users results in random resource availabilities at the secondary networks, which makes the quality-of-service (QoS) support challenging. With the increasing demand for elastic/interactive data services (internet based services) and wireless multimedia services, QoS support becomes essential for CRNs. This research investigates the voice and elastic/interactive data service support over CRNs, in terms of their delay requirements. The packet level requirements of the voice service and session level delay requirements of the elastic/interactive data services are studied. In particular, constant-rate and on-off voice traffic capacities are analyzed over CRNs with centralized and distributed network coordination. Some generic channel access schemes are considered as the coordination mechanism, and call admission control algorithms are developed for non-fully-connected CRNs. Advantage of supporting voice traffic flows with different delay requirements in the same network is also discussed. The mean response time of the elastic data traffic over a centralized CRN is studied, considering the shortest processor time with and without preemption and shortest remaining processor time service disciplines, in comparison with the processor sharing service discipline. Effects of the traffic load at the base station and file length (service time requirement) distribution on the mean response time are discussed. Finally, the relationship between the mean response times of interactive and elastic data traffic is studied.

The growing interest towards wireless communication services over the recent years has increased the demand for radio spectrum. Inefficient spectrum management together with the scarcity of the radio spectrum is a limiting factor for the development of modern wireless networks. As a solution, the idea of cognitive radio networks (CRNs) is introduced to use licensed spectrum for the benefit of the unlicensed secondary users. However, the preemptive priority of the licensed users results in random resource availabilities at the secondary networks, which makes the quality-of-service (QoS) support challenging. With the increasing demand for elastic/interactive data services (internet based services) and wireless multimedia services, QoS support becomes essential for CRNs. This research investigates the voice and elastic/interactive data service support over CRNs, in terms of their delay requirements. The packet level requirements of the voice service and session level delay requirements of the elastic/interactive data services are studied. In particular, constant-rate and on-off voice traffic capacities are analyzed over CRNs with centralized and distributed network coordination. Some generic channel access schemes are considered as the coordination mechanism, and call admission control algorithms are developed for non-fully-connected CRNs. Advantage of supporting voice traffic flows with different delay requirements in the same network is also discussed. The mean response time of the elastic data traffic over a centralized CRN is studied, considering the shortest processor time with and without preemption and shortest remaining processor time service disciplines, in comparison with the processor sharing service discipline. Effects of the traffic load at the base station and file length (service time requirement) distribution on the mean response time are discussed. Finally, the relationship between the mean response times of interactive and elastic data traffic is studied.

Thesis (MScEng)--Stellenbosch University, 2012. / ENGLISH ABSTRACT: In this thesis, the investigation into the optimization of low speed communication protocols for narrow band networks will be presented. The main focus will be on analysing commonly used low speed communication protocols and investigate an alternative to these protocols to provide an optimized low speed narrow band network that provides better performance at high and low channel utilization. A study of existing low speed communication networks within the field of water supply has been used to identify the most appropriate protocols to be included in the analysis. The analysis of each protocol discussed includes the development of a simulation and theoretical model, with parameters based on those obtained from implemented communication networks focusing on the parameters used within the Namib water supply scheme of NamWater in Namibia. Of the currently implemented contention protocols, the non-persistent Carrier Sense Multiple Access (CSMA) protocol is implemented the most. Current models used for modelling these protocols make use of various assumptions. These models have been expanded to provide a more accurate representation of the non-persistent CSMA model. The Round Robin Polling (RRP) protocol is another well known protocol used within the telemetry industry and has also been modelled as an alternative to the non-persistent CSMA model. The Adaptive Tree Walk (ATW) protocol has been identified as the limited contention protocol to be modelled as a possible alternative to the conventional methods used. A new model has been developed for modelling this protocol by making use of the same strategies and tools used in the modelling of the non-persistent CSMA and RRP protocols. The Simulation modelling has been developed by making use of DESMO-J, an Object Orientated Simulation API based in Java, developed by the Faculty of Informatics at the University of Hamburg. DESMO-J has been chosen as an alternative to the more traditional simulation languages due to its complete documentation, support structures, ease of use and flexibility. All theoretical models have been implemented in Matlab. / AFRIKAANSE OPSOMMING: In hierdie tesis sal die ondersoek na die optimering van laespoed protokolle vir kommunikasie oor nouband netwerke voorgelê word. Die hoof fokus is op die analise van algemene laespoed kommunikasie protokolle en die ondersoek van alternatiewe wat ’n meer optimale laespoed nouband netwerk sal lewer deur beter werkverigting by lae en hoë kanaalverkeer. ’n Studie van praktiese laespoed nouband netwerke in die veld van waterverspreiding word gebruik om die mees algemene protokolle te identifiseer wat in die analise ingesluit moet word. Die analise van die protokolle sluit in teoretiese en simulasie modelle, met parameters soos geidentifiseer uit die studie van ‘n praktiese netwerk, naamlik die Namib waterverspreidingskema van NamWater in Namibië. Die ’Non-persistent Carrier Sense Multiple Access’ en ’Round Robin Polling’ protokolle is geidentifiseer as dié wat meeste geimplementeer word. Die werkverigting van die protokolle is geanaliseer deur gebruik te maak van teoretiese en simulasie modeleringstegnieke. Huidige modelle van die CSMA protokol is gebaseer op sekere aannames. Hierdie aannames word aangepas en verbeter vir implementering van die teoretiese model. Die model word ook verder uitgebrei om beter resultate te lewer oor ’n groter parameterstel. Die ’Adaptive Tree Walk’ protokol is geidentifiseer as ’n moontlike optimale protokol en word gemodeleer en vergelyk teen die CSMA en RRP protokolle se werkverigting. Die simulasiemodelle is ontwikkel deur gebruik te maak van die DESMO-J sagteware, soos ontwikkel as ’n Java program-koppelvlak deur die Universiteit van Hamburg se Fakulteit van Informatika. DESMO-J is gekies as ’n alternatief vir die meer tradisionele simuleringstale omrede goeie dokumentasie, maklike gebruik en buigbaarheid. Alle teoretiese modelering is uitgevoer in Matlab.

The objective of this thesis is to automate the design of round-robin arbiter logic. The resulting arbitration logic is more than 1.8X times faster than the fastest prior state-of-the-art arbitration logic the author could find reported in the literature. The generated arbiter implemented in a single chip is fast enough in 0.25ьm CMOS technology to achieve terabit switching with a single chip computer network switch. Moreover, this arbiter is applicable to crossbar (Xbar) arbitration logic. The generated Xbar, customized according to user specifications, provides multiple communication paths among masters and slaves. As the number of transistors on a single chip increases rapidly, there is a productivity gap between the number of transistors available in a chip and the number of transistors per hour a chip designer designs. One solution to reduce this productivity gap is to increase the use of Silicon Intellectual Property (SIP) cores. However, a SIP core should be customized before being used in a system different than the one for which it was designed. Thus, to reconfigure the SIP core, either an engineer must spend significant effort altering the core by hand or else an enhanced CAD tool can automatically customize the core according to customer specifications. In this thesis, we present SIP generator tools for arbiter and Xbar generation. First, we introduce a Round-robin Arbiter Generator (RAG). The RAG can generate a hierarchical Bus Arbiter (BA) which is faster than all known previous approaches. RAG can also generate a hierarchical Switch Arbiter (SA) which is faster than all known previous approaches. Using a 0.25ьm TSMC standard cell library from LEDA Systems, we show the arbitration time of a 32x32 SA and demonstrate that our SA meets the time constraint to achieve terabit throughput. Furthermore, using a novel token-passing hierarchical arbitration scheme, our 32x32 SA performs better than the Ping-Pong Arbiter and Programmable Priority Encoder by factors of 1.8X and 2.3X, respectively, with less power dissipation. Finally, we present an Xbar switch Generator (X-Gt) tool that automatically configures a crossbar for a multiprocessor System-on-a-Chip (SoC). An Xbar is generated in Register Transfer Level (RTL) Verilog HDL.

L'étude des systèmes en réseau supports d'applications collaboratives, distribuées et interconnectées par un réseau repose sur l'identification des exigences de fonctionnement de l'application appelées Qualité de Contrôle (QdC), et sur l'évaluation des performances du réseau pour obtenir son niveau de Qualité de Service (QdS). « Cette thématique comporte d'importants verrous de nature fondamentale relevant du domaine de l'automatique, de la robotique, des capteurs, de la théorie de l'information, des réseaux. Par souci de simplification, les travaux sur les systèmes en réseau se repartissent selon deux approches: la première compense les perturbations générées par les communications au niveau de l'application (« control over network »). La seconde adapte les performances du réseau en fonction des besoins applicatifs (« control of network »). L'objectif de nos travaux de thèse est donc de proposer un environnement de modélisation intégré permettant de représenter le comportement des SCRs. Nous avons choisi les Réseaux de Petri de haut niveau qui possède un fort pouvoir d'expression, de formalisation et dont la modularité permet d'ajouter et/ou de faire évoluer les modèles qui sont développés dans ce travail. Dans un premier temps, nous avons proposé un modèle Ethernet Commuté gérant des mécanismes d'ordonnancement. Le choix de ce réseau a été guidé par le fait qu'il est de plus en plus utilisé dans les SCRs. Ensuite, le modèle d'un SCR a été proposé, et modélisé par des Réseaux de Petri de haut de niveau, en intégrant au modèle Ethernet Commuté, l'environnement applicatif : Contrôleur, Process,.. Enfin, des stratégies pour commander le réseau de façon à adapter sa Qualité de Service en regard de la Qualité de Contrôle requise par l'application, ont été mises en oeuvre. Pour cela, des ordonnanceurs à priorité stricte et de type WRR sont utilisés. Les résultats de simulation montrent clairement que des dispositifs de compensation du réseau pour améliorer les performances du système de communication, permettent aussi d'améliorer les performances du système à commander.

Scheduling is extremely important for modern real-time systems. It enables several programs to run in parallel and succeed with their tasks. Many systems today are real-time systems, which means that good scheduling is highly needed. This thesis aims to evaluate the real-time scheduling algorithm earliest deadline first, newly introduced into the Linux kernel, and compare it to the already existing real-time scheduling algorithms first in, first out and round robin in the context of firm tasks. By creating a test program that can create pthreads and set their scheduling characteristics, the performance of earliest deadline first can be evaluated and compared to the others. / Schemaläggning är extremt viktigt för dagens realtidssystem. Det tillåter att flera program körs parallellt samtidigt som deras processer inte misslyckas med sina uppgifter. Idag är många system realtidssystem, vilket innebär att det finns ett ytterst stort behov för en bra schemaläggningsalgoritm. Målet med det här examensarbetet är att utvärdera schema-läggningsalgoritmen earliest deadline first som nyligen introducerats i operativsystemet Linux. Målet är även att jämföra algoritmen med två andra schemaläggningsalgoritmer (first in, first out och round robin), vilka redan är väletablerade i Linux kärnan. Det här görs med avseende på processer klassificerade som firm. Genom att skapa ett program som kan skapa pthreads med önskvärda egenskaper kan prestandan av earliest deadline first algoritmen utvärderas, samt jämföras med de andra algoritmerna.

The aim of master thesis was to create a program for the quantitative analysis of data backup schemes and with its help to identify and analyze the properties of commonly used schemes for different loads. Based on the obtained results, compile a set of principles for choosing the optimal data backup scheme. The program was created by Matlab. It can be used to find out parameters for individual backup schemes: Parameters C (total backup volume) and E (medium volume of recovery backups), backups size of individual days, workload of individual storages, cost of storages and cost of overall backup scheme, size the total amount of data written per storage per time slot. At the end of the thesis is defined a scheme for choosing the optimal backup scheme.