Målet med denna rapport är att beskriva hur erbjudandesystemet i kandidatprojektet Kunderbjudande i molnet kan implementeras så att man skapar värde för kunden. Utöver detta behandlas erfarenheter från projektet, vilket stöd man kan få genom att skapa och följa upp en systemanatomi samt vilken effekt verktyget Slack har haft på gruppens kommunikation. Vidare tas följande ämnen upp i individuella bidrag till rapporten: Xamarin test recorder Lättviktig kodgranskning med GitHub pull requests Utveckling av mobilapplikation till flera plattformar i Xamarin Kvalitetsplanens roll i programutveckling Kanban med Trello i kandidatprojekt Reacts koddelning mellan webb och mobilapplikationer Tidsbudgetering och estimering inom projektet Kunden Byggvarulistan AB var i behov av en prototyp till ett system där användare kan sparasina kvitton direkt i mobilen och sedan använda dessa för att ta del av erbjudanden. Metoden för att åstadkomma detta var att utveckla ett system som möjliggör hantering av kvitton. Via detta system skulle det även gå att erbjuda cashbacks och kampanjer tillanvändaren. Under utvecklingen användes Slack för kommunikation och en systemanatomi togs fram och användes. Utvecklingen delades upp i fyra iterationer och baserades på agila arbetsmetoder. Systemet är uppdelat i tre moduler: en mobilapplikation, en molntjänst och ett webbgränssnitt. Mobilapplikationen är utvecklad för Android med verktyget Xamarin. Molntjänsten består av ett webb-API som har utvecklats med webbramverket ASP.NET och en databas som implementerades med ORM-ramverket Entity framework core. Webbgränssnittet utvecklades i JavaScript och biblioteket React. Dessa tre delar samverkar och ger en användare möjlighet att genom applikationen hitta och utnyttja erbjudanden från Byggvarulistan och en administratör möjlighet att modifiera erbjudanden och användardata genom webbgränssnit-tet. Slutsatser från de erfarenheter som har observerats under projektets gång är att det är effektivare att arbeta tillsammans inom teamet än att arbeta individuellt samt att valet av utvecklingsmiljö hade större effekt på utvecklingen än man först trodde. Slutsatser kunde även dras om att skapandet av en systemanatomi var fördelaktigt då alla gruppmedlem-mar fick en gemensam bild över systemet. Till sist så drogs även en slutsats om att Slackunderlättade kommunikationen väsentligt både generellt för hela gruppen och för de olika utvecklingsteamen.

WebRTC, Web Real-Time Communications, är en samlingprotokoll och format för säker realtidskommunikation med inbyggda möjligheter för strömning av ljud- och bild. WebRTC finns idag inbyggt i ledande webbläsare som exempelvis Google Chrome. Tekniken är patent-och licensfri, och det finns flera implementationer som är av öppen källkod. Studien utvärderar två av dessa källkodsimplementationer från Ericsson och Google. Implementationen används sedan som grund för att bygga in en infrastruktur med WebRTC till företagets VISIARCs ramverk Coffee. En exempelapplikation från den färdiga implementationen av WebRTC jämförs sedan gentemot Skype om bild- och ljudkvalitet i videosamtal.

The industry shows a clear need for synthetically generated realistic network traffic. As a possible solution, this thesis proposes a method for generating such data in an automatic and controllable manner. This thesis first examines the characteristics of real network traffic and analyzes the length of ON/OFF periods. The theory that network traffic exhibits self-similarity and high variability is once again tested and proven, thereby also the fact that the ON/OFF periods of real network traffic comes from a heavy-tailed distribution. Thereafter, the thesis proposes a way to simulate user interaction with real world applications by using a UI testing framework called WinAppDriver. This tool is then used to synthetically generate network traffic, of which the characteristics are analyzed and compared to that of real network traffic. The results show that the generated network traffic is indeed statistically similar to real network traffic. Finally, everything is combined by setting up a whole network of virtual machines with simulated users.

Our society is becoming more digitalized for each day. Now, we are able to gather data from individual users with higher resolution than ever. With the increased amount of data on an individual user level, we can analyze their behavior. This is of interest in many different domains, for example service providers wanting to improve their service for their customers. If they know how their service is used, they have more insight in how they can improve. But, it also imposes additional difficulties. When we reach the individual user, the irregularities in the regular behavior makes it harder to model the normal behavior. In this thesis, we explore data-driven techniques to model and analyze user behaviors. We aim to evaluate existing as well as develop novel technologies to identify approaches that are suitable for use on an individual user level. We use both supervised and unsupervised learning methods to model the user behavior and evaluate the approaches on real world electricity consumption data. Firstly, we analyze household electricity consumption data and investigate the use of regression to model the household's behavior. We identify consumption trends, how data granularity affects modeling, and we show that regression is a viable approach to model user behavior. Secondly, we use clustering analysis to profile individual households in terms of their electricity consumption. We compare two dissimilarity measures, how they affect the clustering analysis, and we investigate how the produced clustering solutions differ. Thirdly, we propose a sequential clustering algorithm to model evolving user behavior. We evaluate the proposed algorithm on electricity consumption data and show how the produced model can be used to identify and trace changes in the user's behavior. The algorithm is robust to evolving behaviors and handles both dynamic and incremental aspects of streaming data.

The OpenModelica compiler currently generates code directly from a syntax tree representation, which leads to inefficient code in several cases. This thesis work introduces a lower-level intermediate representation for the compiler which aims to simplify the compiler back end and enable more optimizations. The resulting design of the representation features flat primitive operations and control flow using basic blocks and terminators. Variables are mutable, unlike SSA-based representations. Introducing the IR did not significantly change the runtime performance of the test programs. The number of lines of code compared to the old back end was reduced to a quarter, this and the simpler representation will help future work on optimization passes and implementing an LLVM-based back end.

Computer science and software engineering education usually contain programming courses that require writing code that is graded. These assignments are corrected through manual code review by teachers or course assistants. The large amount of assignments motivates us to find ways to automatically correct certain parts of the assignments. One method to ensure certain requirements of written code is fulfilled is by using static analysis, which analyzes code without executing it. We utilize Clang-tidy and Clang Static Analyzer, existing static analysis tools for C/C++, and extend their capabilities to automate requirement checking based on existing assignments,such as prohibiting certain language constructs and ensuring certain function signatures match the ones provided in instructions. We evaluate our forked version of the Clang tooling on actual student hand-ins to show that the tool is capable of automating some aspects that would otherwise require manual code review. We were able to find several errors, even in assignments that were considered complete. However, Clang Static Analyzer also failed to find a memory leak, which leads us to conclude that despite the benefits,static analysis is best used as a complement to assist in finding errors.

Mobile Edge Clouds (MECs) are platforms that complement today's centralized clouds by distributing computing and storage capacity across the edge of the network, in Edge Data Centers (EDCs) located in close proximity to end-users. They are particularly attractive because of their potential benefits for the delivery of bandwidth-hungry, latency-critical applications. However, the control of resource allocation and provisioning in MECs is challenging because of the  heterogeneous distributed resource capacity of EDCs as well as the need for flexibility in application deployment and the dynamic nature of mobile users. To realize the potential of MECs, efficient resource management systems that can deal with these challenges must be designed and built. This thesis focuses on two problems. The first relates to the fact that it is unrealistic to expect MECs to become successful based solely on MEC-native applications. Thus, to spur the development of MECs, we investigated the benefits MECs can offer to non-MEC-native applications, i.e., applications not specifically engineered for MECs. One class of popular applications that may benefit strongly from deployment on MECs are cloud-native applications, particularly microservice-based applications with high deployment flexibility. We therefore quantified the performance of cloud-native applications deployed using resources from both cloud datacenters and edge locations. We also developed a network communication profiling tool to identify the aspects of these applications that reduce the benefits they derive from deployment on MECs, and proposed design improvements that would allow such applications to better exploit MECs' capabilities. The second problem examined in this thesis relates to the dynamic nature of resource demand in MECs. To overcome the challenges arising from this dynamicity, we make use of statistical time series models and machine learning techniques to develop two workload prediction models for EDCs that account for both user mobility and the correlation of workload changes among EDCs in close physical proximity.

Cloud computing offers pay-per-use on-demand access to computer resources for hosting program execution environments for software service deployment. Management of cloud resources includes determining, based on current monitored resource availability, which part(s) of a computational infrastructure should host such program execution environments in a process called placement. Our work defines directives that lets consumers of cloud resources influence placement to express relationships between cloud services (orchestration) and deployment constraints to uphold for related service components, without surrendering the ultimate control over placement from the infrastructure owner. The infrastructure owner remains free to define their policies and placement optimization criteria, e.g., to consolidate work that needs to be done to as few physical host machines as possible for power savings reasons. We show how the placement process can be adjusted to take such influence into account and validate through simulations that the adjustments produce the correct result without too large computational impact on the placement process itself. Further, we present a technique for transferring large data files between cloud data centers that operate in (separate) cloud federations that avoids repeated transfers in a delegation chain between members of (different) cloud federations. Finally, we present a non-invasive method of extracting monitoring data from a service deployed in a cloud federation, and a framework for making monitoring information available and understandable in spite of technical differences between monitoring systems used in cloud federations.

Den här rapporten behandlar ett kandidatarbete som utfördes av åtta stycken studenter från civilingenjörsutbildningarna datateknik och mjukvaruteknik på Tekniska Högskolanvid Linköpings Universitet. Målet med projektet var att utveckla en applikation för att visualisera kontinuerlig integration. Beställning av applikationen gjordes av kunderna Ola Leifler och Kristian Sandahl på Institutionen för Datateknik (IDA) på Linköpings universitet å Software Centers vägnar. Utvecklingsarbetet utfördes enligt den agila utvecklingsmetodiken Scrum med diverseanpassningar. Under utvecklingsarbetets gång gjordes olika sorters prototyper för att säkerställa vilka krav kunden hade på applikationen samt att projektgruppens och kundenstankar om applikationen överensstämde. Projektet resulterade i en applikation som visualiserar kontinuerlig integration i tre olika nivåer. Projektgruppen erhöll även erfarenheter inom utveckling av mjukvara från start till leverans, hur man reder ut en kunds krav med hjälp av prototyper samt gruppdynamik i en projektgrupp. Rapporten innehåller åtta stycken individuella bidrag där varje projektmedlem har skrivit en rapportdel om en erfarenhet eller fördjupning inom ett område kopplad till sin projektroll eller utvecklingen.

Software testing is a complex activity based on reasoning, decision making, abstraction and collaboration performed by allocating multiple cognitive resources. While studies of behaviour and cognitive aspects are emerging and rising in software engineering research, the study of cognition in software testing has yet to find its place in the research literature. Thus, this study explores the cognitive processes used by humans engaged in the software testing process by evaluating an existing: the Test Design Cognitive Model (TDCM). To achieve our objective, we conducted an experiment with veparticipants. The subjects were asked to test a specic Java program and verbalizing their thoughts. The results were analyzed using the verbal protocol analysis guidelines. The results suggest thatall the participants followed similar patterns as those hypothesised by the TDCM model. However,certain patterns specic to each participant were also evident. Two participants exhibited a failure pattern, in which both followed the same sequence of actions after evaluating the test case and verifying that it fails. The experiment protocol and the preliminary results provide a framework from which to investigate tester knowledge and expertise and it is the first step in understanding,evaluating and rening the cognitive processes of software testing.