Local-first software has the potential to offer users a great experience by combining the best aspects of traditional applications with cloud-based applications. However, not much has been documented regarding developing backends for local-first software, particularly one that is scalable while still supporting end-to-end encryption. This thesis presents a backend architecture that was then implemented and evaluated. The implementation was shown to be scalable and was able to maintain an estimated end-to-end latency of around 30-50ms as the number of simulated clients increased. The architecture supports end-to-end encryption to offer user privacy and to ensure that neither cloud nor service providers can access user data. Furthermore, by occasionally performing snapshots the encryption overhead was shown to be manageable compared to the raw data, at around 18.2% in the best case and 118.9% when using data from automerge-perf, a standard benchmark. Lastly, the processing times were shown to be upwards of 50 times faster when using snapshots compared to handling individual changes.

Emerging markets, development of new technologies, sustainability policies and change in customer demand are dramatically changing today’s economies. The business environment is more dynamic than ever before. One particular industry that is currently influenced by significant transformational forces is the automotive industry. These transformational forces, such as autonomy, connectivity and electrification, are driving the change towards a future state where shorter lead times and closer customer interaction will be essential in order to satisfy the needs of the changing market. To deal with external changes, organizations are increasingly focusing on agility as a way to gain new forms of competitive advantage. The benefits of agile methods at the team level has inspired the use of agile practices at a larger scale, all the way up to the portfolio level. Project portfolio management (PPM) connects the strategy of the organization with the distribution of resources across projects in the portfolio. There is little empirical evidence on the agile methods performed in PPM, thus of interest to study. Due to the high complexity at the portfolio level, it becomes increasingly difficult to scale agile methods. This thesis is based on a single-case study within the automotive industry to gain a better understanding of how a mature automotive company can manage the agile scalability to become more agile in their project portfolio management. An abductive approach was applied with gathering methods comprising interviews, documents and observations. The results of this thesis highlight the major deficiencies with the case company’s current PPM, as well as their performance in scaling agile. / Tillväxtmarknader, utveckling av ny teknik, hållbarhetspolitik samt förändring av kundkrav påverkar dagens ekonomi dramatiskt. Affärsmarknaden är mer dynamisk än någonsin tidigare. En viss industri som för närvarande påverkas av bemärkta förändringar är fordonsindustrin. Trender som autonomi, elektrifiering och uppkopplade fordon driver förändringen mot ett framtida tillstånd där kortare ledtider och närmare kundinteraktion kommer att vara nödvändiga för att tillgodose marknadsbehoven. För att hantera externa förändringar fokuserar organisationer alltmer på att arbeta agilt. Fördelarna som agila metoder har på teamnivå har inspirerat användningen av dem på en större skala, hela vägen upp till portföljnivå. Projektportföljstyrning (PPM) förenar organisationens strategi med fördelningen av resurser på projekt i portföljen. Det existerar lite empiriskt bevis på användandet av agila metoder på portföljnivå, därav intresse att studera. Den höga komplexitet som portföljnivån utgör gör det svårare att skala upp agila arbetssätt. Denna avhandling bygger på en fallstudie inom fordonsindustrin för att skapa en bättre förståelse för hur ett moget fordonsföretag hanterar den agila skalbarheten för att bli mer agila i sin portföljstyrning. En abduktiv metod användes med datainsamlingsmetoder som omfattade intervjuer, dokument och observationer. Resultatet av denna avhandling belyser bristerna med företagets nuvarande PPM, samt deras prestanda i agil skalbarhet.

<p>  </p> <p>Federated learning (FL) has emerged as a promising approach for training machine learning models on distributed data while ensuring privacy preservation and data locality. However, one key challenge in FL optimization is the lack of high probability guarantees, which can undermine the trustworthiness of FL solutions. To address this critical issue, we introduce Federated Averaging with post-optimization (FedAvg-PO) method, a modification to the Federated Averaging (FedAvg) algorithm. The proposed algorithm applies a post-optimization phase to evaluate a short list of solutions generated by several independent runs of the FedAvg method. These modifications allow to significantly improve the large-deviation properties of FedAvg which improve the reliability and robustness of the optimization process. The novel complexity analysis shows that FedAvg-PO can compute accurate and statistically guaranteed solutions in the federated learning context. Our result further relaxes the restrictive assumptions in FL theory by developing new technical tools which may be of independent interest. The insights provided by the computational requirements analysis contribute to the understanding of the scalability and efficiency of the algorithm, guiding its practical implementation.</p>

On the web, data visualisation through charts and diagrams can help present data in a more readable way. This is often done through the usage of JavaScript libraries. We experimented with 5 JavaScript data visualisation libraries to determine their respective performances and how each one scaled with increased data size. Our results will hopefully provide help with the selection of said libraries. The results show a significant difference in response times between all libraries for mostdata sizes, with only a few exceptions. Different exponential growths were also identified for all libraries, and the performance often varied greatly depending on chart type. Response time is not the only variable in performance measurements. Future research could cover other aspects, like memory consumption and rendering requirements. There were also times when the libraries did not render at larger data sizes, despite showing no errors, and further investigation behind this should be done.

This is a comparative study of open-source IoT middleware platforms with the main focus on scalability and reliability. An initial evaluation of available open-source IoT platforms resulted in Kaa and Node-RED being the focus of this thesis. To further analyse the platforms, they were both subjected to testing with three real-world scenarios. The chosen scenarios were a remote-controlled LED, a chat application and a data transmitting sensor. Prototypes were developed for each scenario using a range of programming languages and devices like Raspberry Pi, Android and ESP8266.According to the tests Node-RED has better performance on a single server. It also scales better with the possibility to communicate with external APIs directly unlike Kaa which would require a gateway. Despite these factors, Kaa proved to have better overall scalability and reliability with its built-insecurity and device discovery, it also supports clustering and should prove better in larger environments. / Detta arbete är en komparativ studie av olika open-source IoT plattformarmed inriktning på skalbarhet och pålitlighet. I förstudien som utfördes undersöktes tillgängliga open-source plattformar varefter Kaa och Node-RED valdes som lämpliga studieobjekt. För att ytterligare analyseraplattformarna testades de med hjälp av tre olika verklighetsbaseradescenarion. De utvalda scenariona som användes var en fjärrstyrd LED, en chatapplikation och data som skickas trådlöst från en sensor. Prototyperna utvecklades för varje scenario med hjälp av flera olika programmeringsspråksamt olika hårdvaror som Raspberry Pi, Android och ESP8266.Enligt de tester som genomförts så har Node-RED bättre prestandaunderanvändning på en ensam server. Node-RED har även bättre skalbarhet motexterna API:er då detta endast är möjligt i Kaa via en gateway. Trots detta så tyder resultaten på att Kaa sammantaget har bättre skalbarhet och tillförlitlighet med plattformens inbyggda säkerhet och dess möjlighet att automatiskt upptäcka enheter. Kaa stödjer även klustring vilket förmodligen presterar bättre i utbredda miljöer.

The Quality of Service (QoS) is a profound concept which is gaining increasing attention in the Internet industry. Best-effort applications are now no longer acceptable in certain situations needing high bandwidth provisioning, low loss and streaming of multimedia applications. New emerging multimedia applications are requiring new levels of quality of services beyond those supported by best-effort networks. Quality of service routing is an essential part in any QoS architecture in communication networks. QoS routing aims to select a path among the many possible choices that has sufficient resources to accommodate the QoS requirements. QoS routing can significantly improve the network performance due to its awareness of the network QoS state. Most QoS routing algorithms require maintenance of the global network¿s state information to make routing decisions. Global state information needs to be periodically exchanged among routers since the efficiency of a routing algorithm depends on link-state information accuracy. However, most QoS routing algorithms suffer from scalability due to the high communication overhead and the high computation effort associated with maintaining accurate link state information and distributing global state information to each node in the network. The ultimate goal of this thesis is to contribute towards enhancing the scalability of QoS routing algorithms. Towards this goal, the thesis is focused on Localised QoS routing algorithms proposed to overcome the problems of using global network state information. Using such an approach, the source node makes routing decisions based on the local state information for each node in the path. Localised QoS routing algorithms avoid the problems associated in the global network state, like high communication and processing overheads. In Localised QoS routing algorithms each source node maintains a predetermined set of candidate paths for each destination and avoids the problems associated with the maintenance of a global network state by using locally collected flow statistics and flow blocking probabilities. / Libya's higher education

Titel: Från avfall till resurs: Hinder och möjligheter för uppskalning av cirkulära affärsmodeller i klädindustrin Bakgrund: Dagens resursutnyttjande är inte hållbart och det ökade fokuset kring cirkulära lösningar uppmanar till både behov och intresse för förändring. Dessa anpassningar handlar bland annat om att övergå från en linjär till en cirkulär affärsmodell för att göra bättre nyttjande våra resurser. Cirkulära affärsmodeller blir allt viktigare, men hur dessa kan skalas upp finns utrymme att vidare studera. Syfte: Syftet med denna studie är att identifiera centrala faktorer som har inverkan påskalbarheten av cirkulära affärsmodeller i klädindustrin. Metod: Studien använder en kvalitativ forskningsstrategi med en fallstudiedesign. Forskningsperspektivet i studien är hermeneutiskt och har kompletterats med en abduktiv ansats för att vidareutveckla det hermeneutiska tillvägagångssättet och skapa förståelse. Empirin i studien baseras på nio semistrukturerade intervjuer med både företag och experter från olika multinationella företag. Slutligen används en tematisk analys som analysmetod. Slutsats: Studien har identifierat sex hinder och sex möjliggörare för att skala upp den cirkulära affärsmodellen i klädindustrin. Dessa faktorer identifierades utifrån en analysmodell som byggde på strukturell, strategisk och operativ nivå där vissa faktorer var mer centrala än andra. De hindrenmed den mest avgörande betydelsen för uppskalning av cirkulära affärsmodellen var; eftersläpande och bristfälliga regelverk, komplexa värdekedjor och resurstillgänglighet. De möjliggörarna med den mest avgörande betydelsen var; proaktiva regelverk, samarbete och partnerskap och resurseffektivisering. / Title: From waste to resource: Barriers and enablers for scaling up circular business models in the clothing industry Background: The current use of resources is not sustainable and the increased focus on circular solutions is prompting both the need and interest in change. These adaptations include moving from a linear to a circular business model in order to make better use of our resources. Circular business models are becoming increasingly important, but there is room to study how these can reach scalability. Purpose: The purpose of this study is to identify key factors influencing the scalability of circular business models in the fashion industry. Method: The study utilizes a qualitative research strategy with a case study design. The research perspective in the study is hermeneutic and has been complemented with an abductive approach to further develop the hermeneutic approach and create a modified understanding. The empirical data in the study is based on nine semi-structured interviews with both companies and experts from various multinational corporations. A thematic analysis is employed as the analytical method. Conclusion: The study has identified six barriers and six enablers for scaling up the circular business model in the clothing industry. These factors were identified using an analytical model based on structural, strategic and operational levels, with some factors being more central than others. The barriers with the most crucial importance for scaling up the circular business model were; lagging and inadequate regulations, complex value chains and resource availability. The enablers with the most crucial importance were; proactive regulations, collaboration and partnerships, and resource efficiency. Keywords: Circular business model, resource management, clothing industry, scalability.

A Cognitive Radio Network (CRN) is a technology that allows unlicensed users to utilise licensed spectrum by detecting an idle band through sensing. How- ever, most research studies on CRNs have been carried out without considering the impact of sensing on the performance and security of CRNs. Sensing is essential for secondary users (SUs) to get hold of free band without interfering with the signal generated by primary users (PUs). However, excessive sensing time for the detection of free spectrum for SUs as well as extended periods of CRNs in an insecure state have adverse effects on network performance. Moreover, a CRN is very vulnerable to attacks as a result of its wireless nature and other unique characteristics such as spectrum sensing and sharing. These attacks may attempt to eavesdrop or modify the contents of packets being transmitted and they could also deny legitimate users the opportunity to use the band, leading to underutilization of the spectrum space. In this context, it is often challenging to differentiate between networks under Denial of Service (DoS) attacks from those networks experiencing congestion. This thesis employs a novel Stochastic Activity Network (SAN) model as an effective analytic tool to represent and study sensing vs performance vs security trade-offs in CRNs. Specifically, an investigation is carried out focusing on sensing vs security vs performance trade-offs, leading to the optimization of the spectrum band’s usage. Moreover, consideration is given either when a CRN experiencing congestion and or it is under attack. Consequently, the data delivery ratio (PDR) is employed to determine if the network is under DoS attack or experiencing congestion. In this context, packet loss probability, queue length and throughput of the transmitter are often used to measure the PDR with reference to interarrival times of PUs. Furthermore, this thesis takes into consideration the impact of scalability on the performance of the CRN. Due to the unpredictable nature of PUsactivities on the spectrum, it is imperative for SUs to swiftly utilize the band as soon as it becomes available. Unfortunately, the CRN models proposed in literature are static and unable to respond effectively to changes in service demands. To this end, a numerical simulation experiment is carried out to determine the impact of scalability towards the enhancement of nodal CRN sensing, security and performance. Atthe instant the band becomes idle and there are requests by SUs waiting for encryption and transmission, additional resources are dynamically released in order to largely utilize the spectrum space before the reappearance of PUs. These additional resources make the same service provision, such as encryption and intrusion detection, as the initial resources. To this end,SAN model is proposed in order to investigate the impact of scalability on the performance of CRN. Typical numerical simulation experiments are carried out, based on the application of the Mobius Petri Net Package to determine the performance of scalable CRNs (SCRNs) in comparison with unscalable CRNs (UCRNs) and associated interpretations are made.

No description available.

No description available.