Rapporten är en sammanställning av gruppens samlade erfarenheter och lärdomar av ett projekt vars syfte var att utveckla ett verktyg som ska visualisera svarstider mellan mobila klienter och datacenter. Genom att använda kontinuerlig prototypning har gruppen kunnat arbeta på ett användarna sätt för att uppfylla kundens verkliga behov. För att uppnå detta användes utvecklingsmetodiken Kanban. Under projektets gång anpassades metodiken för att bättra passa in i arbetet. Projektets användartester har lett till sammanfattande erfarenheter om visualisering av data. Visualiseringar som projektgruppen ansetts tydliga uppfattades inte alltid på samma sätt av användarna. Att visualisera flera parametrar på en världskarta anses vara problematiskt då kartan i sig endast består av länder. För visualisering av flera parametrar måste då även externa användas, exempelvis cirklar eller andra former.

The aim of the project was to implement a machine learning model to optimise the power consumption of Ericsson’s Kista data center. The approach taken was to use a Reinforcement Learning agent trained in a simulation environment based on data specific to the data center. In this manner, the machine learning model could find interactions between parameters, both general and site specific in ways that a sophisticated algorithm designed by a human never could. In this work it was found that a neural network can effectively mimic a real data center and that the Reinforcement Learning policy "TD3" could, within the simulated environment, consistently and convincingly outperform the control policy currently in use at Ericsson’s Kista data center.

The increasing popularity and improvements in the blockchain technology that offers decentralized communication and transactions in the form of cryptocurrencies that have now days a market value of almost 1.3 Trillion dollars and a huge potential to contribute to other fields such as health care, financial transactions, information technology, secure data exchange, data storage and many others has been pushing towards more integrating of renewable energies in this field. The process of approving and inserting the information and contracts with Bitcoins on the blockchain is called mining of Bitcoin and accounts for a large electricity consumption which has been estimated to be around 120 Terawatt_hour (TWH) worldwide in 2023. To keep this field of technology improving and strengthen its development, leading it towards more usage of clean energy will benefit the field and most importantly will help our societies to face climate change and align with the United Nations sustainable goals regarding integrating and increasing the usage of renewable energies and sustainable methods in all fields of life and industry.  The purpose of this study is to assess the feasibility of building a mining farm for bitcoin powered by renewable energies, solar and wind, by using photovoltaics systems and wind turbines with the integration of storage mediums to utilize the clean energy as much as possible. The project is done by firstly conducting a literature review about the technologies used followed by choosing the most appropriate alternatives that fit best for the current project properties and goals. After choosing the components and methods to be used , the technical feasibility is analyzed by simulating the hybrid energy system using a code program written in the software MATLAB which optimizes and calculates on one hand the electricity production of the system that is used for the hourly reliability in meeting the load demand of the mining devices and on the other hand the total cost of the system which will be built upon to estimate the levelized cost of energy and hence analyze the economic feasibility of the project.After conducting the simulation and financial calculations, the results show that the project is technically feasible and the reliability can be as high as around 8650 hours of the year, however, to achieve a 100 percent reliability a support power should be used such as a diesel generator which has also been done in this work. The economic feasibility indicates that the project will be profitable based on the installed capacities and mined Bitcoins, however a storage medium and a support power production source are vital for the success of such a project.

More and more people are using the internet for data processing, transfer, and storage. With it comes a higher demand for computational power from data servers. Unsurprisingly, the data center industry is becoming an increasingly large industry that is important for people’s daily lives. Data centers cover 2 % of the world’s total electrical consumption and this number is expected to become higher. Running data centers with optimal performance while operating efficiently and as sustainably as possible is a task that is of utmost importance.The way data centers are cooled today is through a CRAH unit that features cooling coils and a fan, the fan blows air over cold coils to prevent damage to server components. Another task for this fan is to create a high differential pressure over the servers using this air, to ensure the air flows in the right direction. The air is uniformly distributed over the servers. With dynamic air-handling measures, it is possible to match the cooling for individual servers, because all servers have different workloads. They generate different amounts of heat. This thesis investigates manual redistribution between servers and how an air-handling damper unit, that sits on the server, is designed to investigate how it can reduce total power draw. Different tests are run in a wind tunnel which houses room for six servers whereas three prototypes are mounted on three of the servers. The main idea to test is that instead of running an even amount of stress on six servers, the same amount of stress is redistributed on only three servers. The ones now running idle have a damper unit blocking the server's rear side. That way the CRAH fan is using less power to create the same differential pressure. Also, the total power draw to all servers is reduced as well. One of the tests was the conventional way of cooling servers today and it had a total power draw of 1362 watts. The test with both redistribution, dampers closed at the rear and turned off servers had a power draw of 951 watts. That is a 30% decrease.

The notions of "network embedded cloud", also known as a "network enabled cloud" or a "carrier cloud", is an emerging technology trend aiming to integrate network services while exploiting the on-demand nature of the cloud paradigm. A network embedded cloud is a distributed cloud environment where data centers are distributed at the edge of the operator's network. Distributing data centers or computing resources across the network introduces topological and geographical locality dependency. In the case of a network enabled cloud, in addition to the information regarding available processing, memory, and storage capacity, resource management requires information regarding the network's topology and available bandwidth on the links connecting the different nodes of the distributed cloud. This thesis project designed, implemented, and evaluated the use of open shortest path first with traffic engineering (OSPF-TE) for propagating the resource status in a network enabled cloud. The information carried over OSPF-TE are used for network-aware scheduling of virtual machines. In particular, OSPF-TE was extended to convey virtualization and processing related information to all the nodes in the network enabled cloud. Modeling, emulation, and analysis shows the proposed solution can provide the required data to a cloud management system by sending a data center's resources information in the form of new opaque link-state advertisement with a minimum interval of 5 seconds. In this case, each embedded data centers injects a maximum 38.4 bytes per second of additional traffic in to the network.<p> / Ett "network embedded cloud", även känt som ett "network enabled cloud" eller ett "carrier cloud", är en ny teknik trend som syftar till att tillhandahålla nätverkstjänster medan on-demand egenskapen av moln-paradigmet utnyttjas.  Traditionella telekommunikationsapplikationer bygger ofta på en distributed service model och kan använda ett "network enabled cloud" som dess exekverande plattform. Dock kommer sådana inbäddade servrar av naturliga skäl vara geografiskt utspridda, varför de är beroende av topologisk och geografisk lokalisering. Detta ändrar på resurshanteringsproblemet jämfört med resurshantering i datacentrum. I de fall med ett network enabled cloud, utöver informationen om tillgängliga CPU, RAM och lagring, behöver resursfördelningsfunktionen information om nätverkets topologi och tillgänglig bandbredd på länkarna som förbinder de olika noderna i det distribuerade molnet. Detta examensarbete har utformat, tillämpat och utvärderat ett experiment-orienterad undersökning av användningen av open shortest path first med traffich engineering (OSPF-TE) för resurshantering i det network enabled cloud. I synnerhet utvidgades OSPF-TE till att förmedla virtualisering och behandla relaterad information till alla noder i nätverket. Detta examensarbete utvärderar genomförbarheten och lämpligheten av denna metod, dess flexibilitet och prestanda. Analysen visade att den föreslagna lösningen kan förse nödvändiga uppgifter till cloud management system genom att skicka ett datacenters resursinformation i form av ny opaque LSA (kallat Cloud LSA) med ett minimumintervall av 5 sekunder och maximal nätverksbelastning av 38,4 byte per sekund per inbäddade data center.

Les centres de données deviennent de plus en plus importants, allant de petites fermes de serveurs distribuées à des grandes fermes dédiées à des tâches spécifiques. La diffusion de services "dans le nuage" conduit à une augmentation incessante de la demande de trafic dans les centres de données. Dans cette thèse, nous étudions l'évolution des réseaux dans les centres de données et proposons des solutions à court et à long terme pour leur intra-connexion physique. Aujourd'hui, la croissance de la demande de trafic met en lumière la nécessité urgente d’interfaces à grande vitesse capables de faire face à la bande passante exigeant de nouvelles applications. Ainsi, à court terme, nous proposons de nouveaux transpondeurs optiques à haut débit, mais à faible coût, permettant la transmission de 200 Gb /s utilisant des schémas de modulation en intensité et à détection directe. Plusieurs types de modulations d’impulsions en amplitude avancées sont explorés, tout en augmentant la vitesse à des débits symboles allant jusqu’à 100 GBd. La génération électrique à haute vitesse est réalisé grâce à un nouveau convertisseur analogique-numérique intégré, capable de doubler les vitesses des entrées et de générer des signaux à plusieurs niveaux d’amplitude. Cependant, le trafic continuera sa croissance. Les centres de données actuels reposent sur plusieurs niveaux de commutateurs électroniques pour construire un réseau d'interconnexion capable de supporter une telle grande quantité de trafic. Dans une telle architecture, la croissance du trafic est directement liée à une augmentation du nombre des composants du réseau, y-compris les commutateurs avec plus de ports, les interfaces et les câbles. Le coût et la consommation d'énergie qui peut être attendus à l'avenir est intenable, ce qui appelle à une réévaluation du réseau. Par conséquent, nous présentons ensuite un nouveau concept fondé sur la commutation de "slots" optiques (Burst Optical Slot Switching, i.e. BOSS) dans lequel les serveurs sont connectés via des nœuds BOSS à travers des anneaux de fibres multiplexé en longueur d'onde et en temps, et organisés dans une topologie en tore. Au cours de cette thèse, nous étudions la mise en œuvre des nœuds BOSS; en particulier, la matrice de commutation et les transpondeurs optiques. L'élément principal au sein de la matrice de commutation est le bloqueur de slots, qui est capable d'effacer n’importe quel paquet (slot) sur n’importe quelle longueur d'onde en quelques nanosecondes seulement. D'une part, nous explorons l'utilisation d'amplificateurs optiques à semi-conducteurs comme portes optiques à utiliser dans le bloqueur des slots, et étudier leur cascade. D'autre part, nous développons un bloqueur de slots intégré monolithiquement capable de gérer jusqu'à seize longueurs d'onde avec la diversité de polarisation. Ensuite, nous présentons plusieurs architectures de transpondeur et nous étudions leur performance. La signalisation des transpondeurs doit répondre à deux exigences principales: le fonctionnement en mode paquet et la résistance au filtrage serré. D'abord, nous utilisons des transpondeurs élastiques qui utilisent des modulations Nyquist N-QAM, et qui adaptent le format de modulation en fonction du nombre de nœuds à traverser. Ensuite, nous proposons l'utilisation du multiplexage par répartition orthogonale de la fréquence en cohérence optique (CO-OFDM). Avec une structure de paquet inhérente et leur grande adaptabilité fréquentielle, nous démontrons que les transpondeurs CO-OFDM offrent une capacité plus élevée et une meilleure portée que leurs homologues Nyquist. Finalement, nous comparons notre solution BOSS avec la topologie Clos replié utilisée aujourd'hui. Nous montrons que notre architecture BOSS nécessite 400 fois moins de transpondeurs et de câbles que les réseaux de commutation électronique d'aujourd'hui, ce qui ouvre la voie à des centres de données hautement évolutifs et durables / Data centers are becoming increasingly important and ubiquitous, ranging from large server farms dedicated to various tasks such as data processing, computing, data storage or the combination thereof, to small distributed server farms. The spread of cloud services is driving a relentless increase of traffic demand in datacenters, which is doubling every 12 to 15 months. Along this thesis we study the evolution of data center networks and present short- and long-term solutions for their physical intra-connection. Today, rapidly-growing traffic in data centers spotlights the urgent need for high-speed low-cost interfaces capable to cope with hungry-bandwidth demanding new applications. Thereby, in the short-term we propose novel high-datarate low-cost optical transceivers enabling up to 200 Gb/s transmission using intensity-modulation and direct-detection schemes. Several advanced pulse amplitude modulation schemes are explored while increasing speeds towards record symbol-rates, as high as 100 GBd. High-speed electrical signaling is enabled by an integrated selector-power digital-to- analog converter, capable of doubling input baud-rates while outputting advance multi-level pulse amplitude modulations. Notwithstanding, data centers’ global traffic will continue increasing incessantly. Current datacenters rely on high-radix all-electronic Ethernet switches to build an interconnecting network capable to pave with such vast amount of traffic. In such architecture, traffic growth directly relates to an increase of networking components, including switches with higher port-count, interfaces and cables. Unsustainable cost and energy consumption that can be expected in the future calls for a network reassessment. Therefore, we subsequently present a novel concept for intra-datacenter networks called burst optical slot switching (BOSS); in which servers are connected via BOSS nodes through wavelength- and time-division multiplexed fiber rings organized in a Torus topology. Along this thesis we investigate on the implementation of BOSS nodes; in particular, the switching fabric and the optical transceivers. The main element within the switching fabric is the slot blocker, which is capable of erasing any packet of any wavelength in a nanosecond time-scale. On the one hand, we explore the use of semiconductor optical amplifiers as means of gating element to be used within the slot blocker and study their cascadability. On the other hand we develop a monolithically integrated slot blocker capable of handling up to sixteen wavelength channels with dual-polarization diversity. Then we present several transceiver architectures and study their performances. Transceivers’ signaling needs to fulfill two main requirements: packet-mode operation, i.e. being capable of recovering few microsecond –long bursts; and resiliency to tight filtering, which occurs when cascading many nodes (e.g. up to 100). First we build packet-mode Nyquist-pulse-shaped N-QAM transceivers, which adapt the modulation format as a function of the number of nodes to traverse. Later we propose the use of coherent-optical orthogonal frequency division multiplexing (CO-OFDM). With inherent packet structure and high spectral tailoring capabilities, we demonstrate that CO-OFDM-based transceivers offer higher capacity and enhanced reach than its Nyquist counterpart. Finally, we compare our BOSS solution to today’s Folded Clos topology, and show that our BOSS architecture requires x400 fewer transponders and cables than today’s electronic switching networks, which paves the way to highly scalable and sustainable datacenters