Le SDN (Software-Defined Networking) émerge comme une nouvelle architecture pour la programmation des réseaux. A l'origine, l'idée du SDN est de déplacer le plan de contrôle à l'extérieur des équipements, et de permettre ainsi un contrôle déporté de l'ensemble depuis une entité logicielle logique nommée "contrôleur". Le principal avantage d'une telle approche est de centraliser donc toute l'intelligence de gestion du réseau dans le contrôleur, qui s'appuie pour cela sur des protocoles standard et assure par ce biais la reprogrammation de la totalité de la partie du réseau sous son contrôle. L'évolution technologique vers le SDN est toujours en cours dans des scénarios de déploiement programmable et flexible des réseaux mobiles. Le NFV (Network Function Virtualization) est le processus de déplacement ou de migration des fonctions réseau d'un équipement dédié de réseau vers des serveurs génériques dans le Cloud. Les SDN et NFV sont deux technologies étroitement liées qui sont souvent utilisées ensemble. Le couplage fort entre les plans de contrôle et de données, ainsi que les limitations en matière de passage à l'échelle et de flexibilité, font que la virtualisation des réseaux mobiles actuels nécessite non seulement l'utilisation du Cloud Computing mais aussi les récentes innovations telles que SDN et NFV pour pouvoir permettre un déploiement à la demande des services réseaux (Network-as-a-Service) aux utilisateurs. Les lignes de recherche globales de cette thèse s'inscrivent dans deux principaux cas d'utilisation. Ces cas d'utilisation, bien qu'appelés de la "prochaine génération de réseaux mobiles", sont le "Telco" et le "Vertical", qui apparaissent ici couplés, les deux étant traditionnellement complètement séparés. Dans les cas d'utilisation de "télécommunications", nous exploitons les avantages de SDN pour avoir un cadre de contrôle flexible pour les réseaux d'auto-organisation (SON) et la division de traitement dynamique des utilisateurs. Dans le cas d'utilisation de "verticale", nous appliquons divers avantages du protocole SDN et OpenFlow pour utiliser efficacement les ressources radio du réseau de backhaul dans le système de communication train-sol. Notre cadre d'étude du SDN, en général, peut être une solution efficace et alternative pour la gestion RAN (Radio Access Network), c'est-à-dire pour des objectives comme l'optimisation des ressources radio, l'optimisation du réseau, la gestion de la mobilité et l'équilibrage de la charge, peuvent être atteint avec ce cadre. Grâce à l'analyse et l'expérimentation concrète des SDN et NFV pour le RAN, nous montrons que les solutions proposées dans ce travail peuvent apporter un faisceau d'avantages évidents aux réseaux mobiles tels que la flexibilité, la programmabilité, la gestion unifiée et la mise en œuvre de nouveaux services / Software-Defined Networking (SDN) has emerged as a new intelligent architecture for network programmability. The primary idea behind SDN is to move the control-plane outside the switches and enable external control of data-plane through a logical software entity called controller. Such approach benefits mobile network management by brining complete intelligence to the logically centralized controller. Network Function Virtualization (NFV) is the process of relocating or migrating network functions from dedicated hardware to generic servers. SDN and NFV are two closely related technologies that are often used together. The traditional mobile network architecture due to its strongest coupling between control and data planes along with limitations in scalability and flexibility requires the usage of cloud computing along with the recent revolutionary approaches in networking such as SDN and NFV to have an architecture that deploys on demand "Network-as-a-Service" for users. The global research focus of this thesis falls in to two main use cases of next generation mobile networks such as Telco and Vertical. In the telco use cases, we exploit the advantages of SDN to have flexible control framework for both Self-Organizing Networks (SON) and dynamic user processing split. In vertical use case, we apply various advantages of SDN and OpenFlow protocol to efficiently utilize the scare radio resources of wireless backhaul network in the train-to-ground communication system. Our SDN framework in general can be an efficient and alternative solution for RAN management i.e. Radio Optimization, Network Optimization, Mobility Management and Load Balancing can be achieved with such framework. Through analysis and experimentation of SDN frameworks for RAN, we shows that the proposed solutions can bring set of advantages to wireless networks such as flexibility, programmability, unified management, and enables new services

No description available.

No description available.

The integration of The Internet of Things and cloud computing has led to the emergenceof new classes of applications ranging from smart healthcare, smart and precision agriculture,smart manufacturing to smart environmental monitoring. The rapid surge in the useof these applications is expected to generate massive amounts of data with differentcharacteristics that are yet not studied. It can be hypothesised that each IoT-enabledapplication may exhibit a diverse range of characteristics that if modelled correctly, maylead to efcient distributed systems. This thesis aims to study the trafc characteristics ofan IoT-enabled healthcare application to build intelligent policies for scalable IoT-cloudsystems by employing the use of workload prediction and load balancing demonstratedon CloudSim Plus platform. The realistic incoming trafc from the SSiO IoT healthcareapplication system is studied, developed and modeled. Workload prediction algorithmsare developed based on ARIMA and SARIMA. The workload prediction algorithms arethen performed and extensively evaluated to select the one with the best performance,which was SARIMA, outperforming ARIMA by 200% on the basis of MAE, RMSE andMAPE. On the basis of the SARIMA prediction for 2 time periods in advance, theload balancing algorithm is preempted to perform horizontal scaling. The results revealthat the load balancer with SARIMA prediction outperform round robin and active loadbalancers for response time and cost by atleast 64% when it comes to worst case scenario.To conclude, a reflection is commented upon about the load balancing for IoT systemsand the directions this could take in the future for a more holistic sustainable approachon real life platforms.

With the emergence of novel cloud applications and their critical latency demand [1], Quick UDP Internet Connection (QUIC) [2] was proposed as a new transport protocol that is promising to reduce the connection establishment overhead while providing security properties similar to Transport Layer Security (TLS) [3]. However, without an efficient task scheduling mechanism, the high cost for encryption and decryption in QUIC can easily lead to load imbalance among multiple Central Processing Unit (CPU) cores and thus cause a high tail latency. In this paper, we proposed a QUIC connection inter-core migration scheme that can dynamically dispatch QUIC connections among CPU cores while keeping the service continuity. We emulated a scenario where the traffic load on two CPU cores is not even and tried to migrate connections from an overloaded CPU to another idle one. The results showed that the load imbalance can be reduced and both of the two CPU cores can be efficiently utilized. Moreover, the throughput can be improved by around 15% with a set of proper parameters thanks to the inter-core connection migration scheme. It was also revealed that the overhead introduced by the connection inter-core migration mechanism is relatively low when applying the migration to a large connection. / Med uppkomsten av nya molnprogram och deras kritiska latens efterfrågan [1], QUIC [2] föreslogs som ett nytt transportprotokoll som lovar att minska anslutningen etableringskostnader samtidigt som de tillhandahåller säkerhetsfastigheter som liknar TLS [3]. Dock utan en effektiv uppgift schemaläggningsmekanism, den höga kostnaden för kryptering och dekryptering i QUIC kan enkelt leda till obalans mellan flera CPU kärnor och orsakar därför en hög svanslatens. I detta dokument föreslog vi a QUIC-anslutning mellan kärnmigrationsschema som dynamiskt kan skickas SNABBA anslutningar mellan CPU kärnor samtidigt som servicekontinuiteten bibehålls. Vi emulerat ett scenario där trafikbelastningen på två CPU kärnor inte är jämn och försökte migrera anslutningar från en överbelastad CPU till en annan inaktiv. De resultaten visade att belastningsobalansen kan reduceras och båda de två CPU: erna kärnor kan utnyttjas effektivt. Dessutom kan genomströmningen förbättras med cirka 15% med en uppsättning korrekta parametrar tack vare anslutningen mellan kärnor migrationsschema. Det avslöjades också att de omkostnader som infördes av anslutning mellan kärnmigrationsmekanismen är relativt låg vid ansökan migreringen till en stor anslutning.

In computational fluid dynamics, we often need to solve complex problems with high precision and efficiency. We propose a three-pronged approach to attain this goal. First, we use the discontinuous Galerkin spectral element method (DG-SEM) for its high accuracy. Second, we use graphics processing units (GPUs) to perform our computations to exploit available parallel computing power. Third, we implement a parallel adaptive mesh refinement (AMR) algorithm to efficiently use our computing power where it is most needed. We present a GPU DG-SEM solver with AMR and dynamic load balancing for the 2D wave equation. The DG-SEM is a higher-order method that splits a domain into elements and represents the solution within these elements as a truncated series of orthogonal polynomials. This approach combines the geometric flexibility of finite-element methods with the exponential convergence of spectral methods. GPUs provide a massively parallel architecture, achieving a higher throughput than traditional CPUs. They are relatively new as a platform in the scientific community, therefore most algorithms need to be adapted to that new architecture. We perform most of our computations in parallel on multiple GPUs. AMR selectively refines elements in the domain where the error is estimated to be higher than a prescribed tolerance, via two mechanisms: p-refinement increases the polynomial order within elements, and h-refinement splits elements into several smaller ones. This provides a higher accuracy in important flow regions and increases capabilities of modeling complex flows, while saving computing power in other parts of the domain. We use the mortar element method to retain the exponential convergence of high-order methods at the non-conforming interfaces created by AMR. We implement a parallel dynamic load balancing algorithm to even out the load imbalance caused by solving problems in parallel over multiple GPUs with AMR. We implement a space-filling curve-based repartitioning algorithm which ensures good locality and small interfaces. While the intense calculations of the high order approach suit the GPU architecture, programming of the highly dynamic adaptive algorithm on GPUs is the most challenging aspect of this work. The resulting solver is tested on up to 64 GPUs on HPC platforms, where it shows good strong and weak scaling characteristics. Several example problems of increasing complexity are performed, showing a reduction in computation time of up to 3× on GPUs vs CPUs, depending on the loading of the GPUs and other user-defined choices of parameters. AMR is shown to improve computation times by an order of magnitude or more.

Mobile networks have rapidly evolved over decades, and have arrived at the fifth generation (5G) system in recent years. From 2019, companies started to bring 5G networks into business, providing higher data rates, lower latency, and larger network capacity to the world. One of the main advancements in 5G network systems is the use of millimetre waves for wireless transmission. This not only grants higher throughput with wide bandwidth but also introduces new challenges to the radio access networks in 5G systems. To tackle the challenges, a new type of network, which is called the Integrated Access and Backhaul (IAB) network, has been proposed as a deployment solution for 5G cellular networks. In this work, we investigate the routing mechanism of an IAB network and propose a novel load balancing algorithm that can be applied to the IAB network for improvement in throughput as well as load distribution. We extend the work from previous studies on IAB networks and evaluate the performance of the proposed algorithm in comparison to previous work. Through radio network simulations, we find that the shortest path routing outperforms a greedy algorithm in terms of path selection in the network and that the proposed algorithm achieves load balancing among the network, combined with an improvement in the user throughput. Finally, we conclude our work and suggest directions for future work in the study of IAB networks. / Mobilnäten har utvecklats snabbt de senaste decennierna och är nu framme vid femte generationens system (5G). Under 2019 började telekomföretag lansera 5G-nätverk, med högre datahastigheter, lägre fördröjningar och högre nätverkskapacitet. Ett av de största framstegen inom 5G-nätverkssystem är användningen av millimetervågor för trådlös överföring. Detta ger högre datahastigheter och större bandbredd, men leder också till nya utmaningar för radioaccessnätverket. För att hantera några av dessa har en ny typ av nätverk, kallat Integrated Access and Backhaul (IAB) föreslagits. I det här arbetet undersöker vi routingmekanismer för ett IAB-nätverk och föreslår en ny lastbalanseringsalgoritm som kan användas för att förbättra såväl genomströmning som lastfördelning. Arbetet är en utvidgning av tidigare studier av IAB-nätverk och jämför prestanda för den nya algoritmen med tidigare förslag. Genom simuleringar har vi funnit att shortest path routing överträffar en ”greedy” algoritm när det gäller vägval i nätverket och att den föreslagna algoritmen uppnår såväl lastbalansering som förbättrad genomströmning. Avslutningsvis ges förlag till fortsatt arbete inom studiet av IAB-nätverk.

I takt med nya innovationer och samhällets fortsatta utveckling och utbredning ökar även elektrifieringen. För att uppnå flera av de miljömål som satts upp av Sveriges riksdag krävs en övergång till ett elsystem byggt på el från förnybara källor. Den ökande elektrifieringen medför inte bara ökade krav på den mängd el som produceras, men även förmågan att transportera den. Den ökade belastningen på elnätet skapar flaskhalsar och kapacitetsbrist i större städer och regioner, vilket gör det svårt att till exempel etablera nya fabriker och bostadsområden vilket bromsar samhällets utveckling. För att lösa denna kapacitetsbrist behövs det svenska stamnätet byggas ut. Detta är dock en relativt långsam process vilket ställer krav på innovativa lösningar som kan avlasta elnätet på kort och lång sikt. En sådan lösning är en ökad efterfrågeflexibilitet, vilket innebär att vissa effektkrävande aktiviteter flyttas till tider på dygnet då elnätet är mindre belastat. Genom att använda sig av smart elnätsteknik kan konsumenter få information om elpriser i realtid och därmed anpassa sin elkonsumtion och dra nytta av lägre elpriser och större tillgång till grön el då topplasten på elnätet är lägre. Syftet med denna studie är att undersöka hur sådana smarta elnätstjänster kan resultera i en ökad efterfrågeflexibilitet i bostäder, och om denna efterfrågeflexibilitet kan hjälpa till att motverka kapacitetsbristen i det svenska elnätet. För att undersöka detta har en litteraturstudie med fokus på beteendeförändringar kopplade till hållbar utveckling samt fyra kvalitativa intervjuer av relevanta aktörer i elbranschen gjorts. Resultaten från litteraturstudien har analyserats i syfte att undersöka huruvida de är applicerbara på smarta elnätstjänster för att uppnå en ökad efterfrågeflexibilitet eller ej. De kvalitativa intervjuerna har summerats och analyserats för att undersöka vilka frågor det råder enighet samt oenighet i bland branschens aktörer. Efter analys samt diskussionen visar studien att smarta elnätstjänster kopplade till bostäders elkonsumtion har potential att uppnå en ökad efterfrågeflexibilitet samt att denna efterfrågeflexibilitet kan bidra till att minska kapacitetsbristen i elnätet. Detta ska dock inte ses som en universell lösning utan en av flera nödvändiga lösningar för att rå på problemet. / As innovation and society continue their growth, electrification is increasing rapidly. To achieve several of the environmental goals set by the Swedish Parliament, a transition to an electricity system based on electricity from renewable sources is also required. The increasing electrification entails not only increased demands on the amount of electricity produced, but also the ability to transport it. Increased load on the electricity grid creates bottlenecks and capacity shortages in larger cities and regions, which makes it difficult to establish, among other things, new factories and residential areas, which slows down the development of society. To solve this capacity shortage, the Swedish electrical grid needs to be expanded. However, this is a relatively slow process, which places demands on innovative solutions that can relieve the electricity grid, both short term and long term. Such a solution is an increased demand flexibility, which means that certain power-demanding activities are moved to times of the day when the grid is less congested. By using smart grid technology, consumers can get information about electricity prices in real time and thereby adjust their electricity consumption and benefit from lower electricity prices and greater access to renewable energy when the top load on the electricity grid is lower. The purpose of this study is to investigate how such smart grid services can result in increased demand flexibility in homes, and whether this demand flexibility can help counteract the capacity shortage in the Swedish grid. To investigate this, a literature study focusing on behavioral changes linked to sustainable development, and four qualitative interviews with relevant actors in the energy industry, have been conducted. The results from the literature study have been analyzed to investigate whether or not they are applicable to smart grid services to achieve increased demand flexibility. The qualitative interviews have been summarized and analyzed to examine what issues there is agreement and disagreement in among the industry's players. After analysis and discussion, the study shows that smart grid services linked to residential electricity consumption have the potential to achieve increased demand flexibility and that this demand flexibility can contribute to reducing capacity shortage. However, this should not be seen as a universal solution but as one of several necessary solutions to tackle the problem.

Databases as a service (DBaaS) transfer the advantages of cloud computing to data management systems, which is important for the big data era. The allocation in a DBaaS system, i.e., the mapping from databases to nodes of the infrastructure, influences performance, utilization, and cost-effectiveness of the system. Modeling databases and the underlying infrastructure as weighted graphs and using graph partitioning and mapping algorithms yields an allocation strategy. However, graph partitioning assumes that individual vertex weights add up (linearly) to partition weights. In reality, performance does usually not scale linearly with the amount of work due to contention on the hardware, on operating system resources, or on DBMS components. To overcome this issue, we propose an allocation strategy based on penalized graph partitioning in this paper. We show how existing algorithms can be modified for graphs with non-linear partition weights, i.e., vertex weights that do not sum up linearly to partition weights. We experimentally evaluate our allocation strategy in a DBaaS system with 1,000 databases on 32 nodes.

Många organisationer världen över är i behov av att förmedla data lagrade i databaser till sina kunder samt ta emot information av kunden som lägger till/tar bort data i databasen. Att kunna ha databastjänsterna tillgängliga åt kunden under olika omständigheter och behandla kundens transaktion mot databasen snabbt är något som gynnar organisationen. Detta är särskilt viktigt för den svenska myndigheten Bolagsverket. För detta ändamål kan flera databaservrar interagera med varandra genom replikering, vilket resulterar i ett serverkluster av flera sammankopplade noder. I arbetet utforskades det om servrar med Postgres kan möta myndighetens behov. Målet med arbetet är att utifrån en skapad sammanställning av vilka områden som önskas utforskas, att demonstrera om Postgres har vad som krävs för att möta områdets kriterier och slutligen utvärdera utfallet av demonstrationen. Arbetet resulterade i tre sammanställda områden, totalt 12 implementationer över samtliga områden sammanlagt samt 6 av dem som mötte de kriterier för det område som behandlades. I arbetet tillämpades för samtliga implementeringar en direkt/indirekt fysisk replikering, som bygger på att en av noderna kan skrivas åt i taget där förändringen replikeras direkt till övriga noder. I samband med implementering användes ett flertal av Postgres tillägg och externa utbyggnader. Det beslutades under arbetets förlopp att endast behandla två av dessa områden fullständigt till förmån för utförlighet. Slutsatsen drogs att för de två områdena som behandlades fullständigt så hade Postgres förmågan att möta kriterierna. Dessutom så drogs slutsatsen att om en implementering avsedd för det tredje området kunde möta de sammanställda kriterierna så har Postgres det som krävs för att möta myndighetens behov. / Many organizations worldwide are in need of conveying data stored in databases to their customers as well as receiving information from the customer who adds/deletes data in the database. Being able to have the database services available to the customer in various circumstances and to process the customers transaction against the database quickly is something that benefits the organization. This is particularly of importance for the authority of the Swedish Companies Registration Office (Bolagsverket). For this purpose, several database servers can interact with each other through replication, resulting in a server cluster of several interconnected nodes. The work explored whether servers with Postgres can meet Bolagsverkets needs. The goal of the work consisted of three parts. To define what areas to explore and related mandatory criterias. Create solutions based on the previous step. And finally demonstrate that Postgres has what it takes to meet the areas criterias. The work resulted in three areas, a total of 12 implementations across all areas combined, in of which 6 of them met the mandatory criterias. In the work, a direct/indirect physical replication was applied in all implementations. In physical replication one of the nodes can be written to at a time, where the change is replicated directly to the other nodes. Within implementation, a number of official and external Postgres tools/extensions were used combined with Postgres base functionality. It was decided during the course of the work to treat only two of these areas in full in favor of extensive work. It was concluded that for the two areas that were fully addressed, Postgres had the ability to meet the criterias. In addition, it was concluded that if an implementation intended for the third area could meet the criterias, then Postgres has what it takes to meet the authority needs.