本篇論文將針對線上遊戲產業之雲端運算服務平台之發展進行研究與探討。從線上遊戲產業發展的現況與趨勢中，看到線上遊戲仍是市場的主流，更是付費機制的成熟產品，但也看到新興遊戲如SNS網頁遊戲、App遊戲的衝擊下，線上遊戲已不再是大型遊戲或是只有大型公司才有能力開發製作的產品，而中小型團隊也開始蜂擁加入開發的行列。 在線上遊戲產品開發流程與產業價值鏈的文獻探討中，看到線上遊戲的開發與營運價值單元，且這些價值單元都是非常符合雲端運算的功能與架構，並再透過詳細解析產品開發流程與營運價值鏈的每個環節中去對應雲端運算架構與功能，分析雲端運算對線上遊戲之開發端及消費端所提昇之價值與效益面，以及線上遊戲之雲端運算服務平台的可能功能與服務模式。 線上遊戲因為具備文件、物件資料量龐大、即時運算需求極高、中小型開發團隊須降低軟硬體開發成本需求殷切以及需要共用之伺服器軟硬體環境，還有備援且不允許因維運、修改、除錯等而可能停機或當機之風險產生等特性，使得在這一波雲端運算的潮流與趨勢中，選擇提供線上遊戲開發與營運之雲端運算平台服務，應是一項可預期的發展機會。 本研究最後透過一家已推出線上遊戲雲端運算服務平台的個案研究，再呼應前面線上遊戲與雲端運算功能與架構的關聯框架及分析，最後提出對個案公司應以大量即時運算為核心能力，提供跨業服務、以核心技術發展技轉與合作營運模式、結合HTML5或Yahoo的YUI等技術，全力搶攻新興連線遊戲市場、研發雲端Server端引擎產品等建議外，也對線上遊戲之雲端運算服務平台提出可發展SaaS、PaaS、IaaS等各別專業及可具備彈性且可發展成機動組合的混合型雲端服務模式之結論與建議等。 / This report is the research and investigation of the cloud computing platform development in online game industry. We find that online game is still the mainstream as well as a developed product under the paid mechanisms through the condition and trend of online game industry. However, because of the impact of SNS website game and App game, the online game no longer just a hug online game or develop by big companies, instead the medium and small groups can develop the online game as well now. We find out that there are some unites in the literature investigated the development process and industrial value chain of online fit in with the function of cloud computing and its framework. In addition, we will do the detailed analysis of production developing process and operation value to correspond to the function and framework of cloud computing, analyze the value and benefit the cloud computer can do to improve the beginning and consumption-side of online game and its possible function and service mode for online game. The online game needs huge quantity of data in document and object, so the need of instant cloud computing is high. For that reason, when develop the software and hardware, the medium and small groups need to reduce the cost. Moreover, they need the server that can use in both new software and hardware, the backup, un-interruptible power supply and the feature of operating normally while doing the maintenance, modification, debug, etc. So, in the trend of current cloud computing, one of the predictable development opportunity is have the cloud computing platform service that can help to develop and operate online game. In the end, the study will take the case study of cloud computing platform that have released online game as example, correspond to the relevant framework and analysis from the previous online game and the function and framework of cloud computing and finally give the suggests and conclusion to the case company, which is taking great amount of instant cloud computing as core ability, provide cross-field services, use core technology development transformation and cooperation service mode, The use of HTML5 or Yahoo's YUI technology put in all effort to earn the new market of online game and research and develop the cloud computer server engine. Other than that, we also bring up the individual professional service that can develop in the cloud computing platform of online game such as SaaS, PaaS, IaaS and the suggestions and conclusions relate to the flexible and assembling-based complex cloud service.

Commercial computer gaming is a large growing industry that already has its major contributions in the entertainment industry of the world. One of the most important among different types of computer games are Real Time Strategy (RTS) based games. RTS games are considered being the major research subject for Artificial Intelligence (AI). But still the performance of AI in these games is poor by human standards due to some fundamental AI problems those require more research to be better solved for the RTS games. There also exist some AI algorithms those can help us solve these AI problems. Anytime- Algorithms (AA) are algorithms those can optimize their memory and time resources and are considered best for the RTS games. We believe that by making AI algorithms anytime we can optimize their behavior to better solve the AI problems. Although many anytime algorithms are available to solve various kinds of AI problems, but according to our research no such study is been done to compare the performances of different anytime algorithms for an AI problem in RTS games. This study will take care of that by building our own research platform specifically design for comparing performances of our selected anytime algorithms for an AI problem. / Address: NaN Mob. +46 - 737 - 40 19 17

Commercial computer gaming is a large growing industry, that already has its major contributions in the entertainment industry of the world. One of the most important among different types of computer games are Real Time Strategy (RTS) based games. RTS games are considered being the major research subject for Artificial Intelligence (AI). But still the performance of AI in these games is poor by human standards because of some broad sets of problems. Some of these problems have been solved with the advent of an open real time research platform, named as ORTS. However there still exist some fundamental AI problems that require more research to be better solved for the RTS games. There also exist some AI algorithms that can help us solve these AI problems. Anytime- Algorithms (AA) are algorithms those can optimize their memory and time resources and are considered best for the RTS games. We believe that by making AI algorithms anytime we can optimize their behavior to better solve the AI problems for the RTS games. Although many anytime algorithms are available to solve various kinds of AI problems, but according to our research no such study is been done to compare the performances of different anytime algorithms for each AI problem in RTS games. This study will take care of that by building our own research platform specifically design for comparing performances of selected anytime algorithms for an AI problem

This thesis studies the open dataplatform Trafiklab, which provides openaccess to data regarding public transportin Sweden. The study is from the perspective ofthird-party developersand deals with the question of how theyvalue creating mechanisms towards themcan be improved. It is based on twodifferent surveys and severalinterviews conducted with third-partydevelopers using Trafiklab. The resultsshow that Trafiklab needs to improvetheir documentation, communication,initial use, and change theirperspective and role towards thedevelopers using Trafiklab. To improveopen data, in general, a greater focus ontransparency rather then transparency issuggested. / Detta examensarbete syftar till att skapa förbättringsförslag till en specifik typ avmjukvara utifrån använderupplevelse. Detta område är inom den öppna data branschen,öppna data innebär att man öppnar upp data så att vem som helst ta del och använda sigav den. Inom kollektivtrafikbranschen distribueras öppen data via en plattform somheter Trafiklab, som ägs gemensamt av branschens parter via en organisation mednamnet Samtrafiken. Det är inom detta område som denna uppsats försöker besvarafrågan hur Trafiklab kan bli mer värdefullt för tredjepartsutvecklare utifrån enmultidisciplinär ansats. Detta kommer att genomföras genom att både försöka ta reda påvad för värde Trafiklab idag har och hur deras nätverk gör värdeskapande möjligt.Detta projekt bestod av tre delprojekt som alla byggde på varandra. Först genomfördeen förstudie att skapa en grund för vidare undersökning presenteras teori kring öppendata och plattform. En plattform är något som skapar en grund som andra aktörer kanbygga på, ett exempel på detta är Apples iPhone där andra aktörer kan byggakomplementerande applikationer. När man har plattformer som tillhandahåller öppendata är det viktig med en balans mellan kontroll och tillgänglighet, något somsammanfattas i teorin kring ”plattform gränsresurser”. Det andra delprojektet varutformades och genomfördes av en enkät, baserat delvis på en redan genomförd enkät.Detta leda till det tredje delprokelt till mer djupgående intervjuer med enkät deltagarnaför att få en mer detaljerad bild av situationen.Resultatet från studien som helhet presenteras i form av en genomgång av Samtrafikenoch Trafiklab följt av en presentation av tredjepartsutvecklares åsikter om Trafiklab.Utvecklarnas kommentarer var i grunden positiva men det fanns områden därförbättringar skulle kunna genomföras. Detta var inom områden som dokumentation,kommunikation, förenkling att börja utveckla och Trafiklabs syn påtredjepartsutvecklare och deras relation. Detta resultat sammanfattas och diskuterassedan med hjälp av de teoretiska begreppen. Där presenteras en kartläggning avTrafiklabs plattforms ekosystem. Detta följs av en mer grundläggande teoretiskdiskussion som avslutas med att ”plattform gränsresurs modellen” appliceras ochförbättringsförslag ges.Uppsatsen avslutas med att föreslå att begreppet öppen data inte enbart betonartillgänglighet men också transparens. Detta gäller både för Trafiklab och för öppen datai Sverige generellt. Några ytterligare förslag till Trafiklab och på fortsatt intressantforsking inom området ges också som avslutning.

近年來平台商業模式席捲全球，Google、Amazon、Facebook、Airbnb、Line等平台領導者們，改變了人與人之間的互動、翻轉現代的生活，並從平台中取得價值、獲得優勢，進行一場以生態系為主的競爭遊戲。此股炫風吹向經濟體中的不同環節，讓企業與企業、產業與產業之間的互動關係發生變化，同時造成許多無法因應平台經營模式與競爭策略的企業快速衰退。因此，在此股潮流之中，舊有企業如何成功轉向平台經營為本研究之重點。 有別於分析成立之際即以平台模式為主的個案公司，本研究結合了價值鏈經營模式、以及平台經營模式的觀點與相關學術理論，採用縱斷面研究方法探討典範企業—台積電30年來經營模式的轉變歷程，分析其如何一步步重塑半導體產業結構、佈局平台策略並成功轉型，成為半導體產業的巨頭之一，並形成台積大同盟的生態系與兩大IDM對手展開全面競爭。 透過本研究之分析與歸納，發現「價值鏈」與「平台」兩者的商業模式要素內涵與經營思維迥異，前者專注本業、著重內部優化，以低成本或差異化為主要的競爭策略；後者則強調擴張營運範圍、槓桿外部資源，共同創造更多價值獲得雙贏局面。而價值鏈模式之企業可從建立內部平台開始、轉變到供應鏈平台，接著發展至雙邊平台的演化方式進行轉型，但不是所有企業都具備成為雙邊平台領導者的潛力，必須確認自身提供的價值是否為整個技術系統所需、以及能否同時為產業中許多企業解決商業問題。當企業成功轉為雙邊平台之際，即可專注於提高平台的網路效應與轉換成本等策略行動上，追求大者恆大及生態系競爭之目的。 / In recent years, the platform business model sweeps the world. Google, Amazon, Facebook, Airbnb, Line and other platform leaders not only changed the way people lived but also obtained values from the platform’s network. These values are key competitive advantages for platform leaders to win the game of business. However, the new type of business model is threatening pipeline business, and some pipelines even rapidly declined after attacking by platforms. Therefore, how pipelines can survive and turn successfully to platform-based business is the main discussion in this research. This study combined the academic theories of two business model: value chain and platform to explore the transformation of business and strategy of TSMC, which is one of the leading semiconductor companies in the world, in the past 30 years. The case study followed by analyzing how TSMC reshaped the semiconductor’s value chain and then adopt the platform strategy for enterprise transformation. Now, TSMC has its own ecosystem that can beat with other two giant IDM competitors. With the in-depth case study of the changing in TSMC’s business model, we had three conclusions. Firstly, we find out that the business model of “value chain” and “platform” has the different connotation and strategic thinking. The former enterprises focus on its own business, pay attention to internal optimization, and take advantage of differentiation or low cost for competition. However, the later ones emphasize the expansion of operating scope, leverage external resources, and co-create with partners for more values to pursuit the win-win situation. Secondly, enterprises with the value chain business model can develop the platform business model by establishing the internal platform at the initial and then expand to supply chain platform and two-sided market platform. But it's worth noting that not all enterprises have the potential to become a platform leader. It must satisfy two prerequisite conditions: (1) the own value is essential within an industry, and (2) can solve the business problems of different companies at the same time. Thirdly, when enterprises successfully become the platform-based business, enlarging network effect and switching cost are two primary actions to pursue and keep platform leading strategy.

近年來，國人追求健康、愛美等因素而運動風氣蔚為盛行，連帶擴大穿戴型裝置產業發展，越來越多廠商加入。而在競爭激烈的全球市場中，不少廠商以平台經濟模式的建立及善用網路外部性的發展為轉型的策略，其中，如何有效率擴大平台規模及提高顧客黏著度尤其重要。本研究著重於以下三點：個案公司之平台商業模式的運作機制、如何創造不同種類之網路外部性及如何管理網路外部性創造更多效益。 本研究採取單一個案研究法，並挑選從GPS大廠成功涉足智慧型穿戴式裝置且近年來積極轉型之台灣國際航電股份有限公司(Garmin Corp.)，經過研究個案公司平台商業模式與網路外部性的創造及管理，歸納本研究之發現：Garmin透過Garmin Sports平台的管理，建立與外部用戶的關係；並藉由穿戴式裝置的租借與其他廠商合作活動帶動運動市場貿易；再教練培訓為中心出發，提供一套專業、完整的課程，建立平台生態圈，擴大平台規模。 本研究之貢獻在於透過Garmin Sports的案例建立一套平台發展與網路外部性的創造與管理之架構，以期了解透過平台網路外部性發展如何提升使用者價值、為使用者帶來更多效益。歸納出三項因素如下： 1. 專業人士與外部廠商 2. 互補品的發展與資料庫的建立 3. 透過多方用戶的選擇與舉辦活動來管理網路外部性 此外，本研究為Garmin Sports平台成立初期，上線時間較短，部分網路外部性的交互影響研究受限。故建議後續研究，可待Garmin Sports平台運作期間較長，其網路外部性的創造與管理及不同種類網路外部性之相互作用亦會較為明顯。 / In recent years, the atmosphere of physical workout becomes prevalent, which results in the expansion of wearable device industry, causing more and more companies join in this industry. Under such a highly competitive global markets, platform transformation gradually becomes a popular strategy. In the research, our main focus lies on what the platform business model is and how to create and manage different kind of network externality. The case study method is applied for this research. We select Garmin Corp, a GPS corporation which successfully develops wearable devices in recent years and has transformed into a new business model as our research subject. With in-depth case study of platform business model and the creation and management of network externality, we conclude our findings as below: Garmin establishes the relationship with external users and cooperates with other companies through the management of the Garmin Sports platform by leasing wearable devices. To expand the scale of the platform, Garmin provides a set of professional training camp and competitions, so as to establish the platform ecosystem. The academic contribution of this study is to create the structure of developing platform and managing network externality, hoping to bring more value and benefits. The factors are as below: 1. Experts and external companies 2. Complementary products and database setup 3. multi-user selection and managing network externality through events In addition, this study is about the beginning of Garmin Sports platform in which on-line time is short and the interaction of network externality is limited. It is recommended that, for the follow-up research, researches should wait for Garmin Sports platform to operate for a longer period of time. Therefore, the creation and management of network externalities and the different types of network interaction will be more obvious.

Le maintien des tissus péri-implantaires sains à un niveau le plus coronaire possible conditionne le succès à long terme des réhabilitations implanto-portées. Dès lors, la mise en place d'une barrière muqueuse protectrice et la préservation de l'os marginal au cours des premières semaines de cicatrisation est indispensable. Le design implantaire influence les remodelages tissulaires intervenant au niveau du col de l'implant lors des phases initiales de cicatrisation. L'objet de ce travail est d'analyser histologiquement chez le Beagle l'influence de deux systèmes implantaires innovants sur les caractéristiques de la muqueuse et la position de l'os marginal après 3 et 12 semaines de cicatrisation. Les résultats de cette étude, mis en relation avec les données de la littérature indiquent que les implants « deux-parts » présentant une connectique cône morse pourraient favoriser la mise en place d'un espace biologique court et limiter la perte osseuse marginale péri-implantaire. Les implants « une part » à col transmuqueux concave seraient quant à eux associés à des valeurs réduites d'espace biologique et autoriseraient une apposition osseuse sur l'épaule de l'implant. Dans la première partie de ce mémoire, nous établissons une synthèse des données histologiques actuelles concernant les caractéristiques et les processus de cicatrisation des tissus péri-implantaires, et l'influence de la conception implantaire sur les remodelages tissulaires marginaux intervenant au cours de la cicatrisation. Nous avons rédigé deux publications internationales que nous présentons en deuxième partie. Une réflexion analytique autour de ce travail est présentée dans la troisième partie / Aesthetic and functional long-term success of implant-supported rehabilitations relies on the preservation of peri-implant tissue levels in the most coronal position. Therefore, the achievement of an efficient protective mucosal seal, and the preservation of the peri-implant marginal bone during the first weeks of healing are essential to prevent long term implant failures. The characteristics of an implant transmucosal design (connections, platforms, surface properties) are related to biological width dimensions, marginal peri-implant bone levels, and the amount of inflammation within the peri-implant soft tissues. The aim of the research work presented in this manuscript was to investigate the effect of two innovative implant systems on peri-implant mucosa maturation, dimensions, collagen fiber organization, and marginal bone levels after 3 and 12 weeks of healing in the beagle dog. The results, compared with previous data, show that platform-switched twopiece implants exhibit reduced values of biological width and marginal bone loss, and that a concave transmucosal design in one-piece implants is associated with a short vertical value of biological width, and promote a mechanical interlocking of the implant body at the connective tissue and marginal bone levels. In the first part of this report, an overview of the current knowledge concerning the characteristics and healing patterns of peri-implant tissues, and the influence of implant design on the early peri-implant tissue remodeling is established. We have written two international publications which are presented in the second part. An analytical reflection about this work is presented in the third part

Coarse-Grained Reconfigurable Architectures(CGRAs) can be employed for accelerating computational workloads that demand both flexibility and performance. CGRAs comprise a set of computation elements interconnected using a network and this interconnection of computation elements is referred to as a reconfigurable fabric. The size of application that can be accommodated on the reconfigurable fabric is limited by the size of instruction buffers associated with each Compute element. When an application cannot be accommodated entirely, application is partitioned such that each of these partitions can be executed on the reconfigurable fabric. These partitions are scheduled by an orchestrator. The orchestrator employs dynamic dataflow execution paradigm. Dynamic dataflow execution paradigm has inherent support for synchronization and helps in exploitation of parallelism that exists across application partitions. In this thesis, we present a compiler that targets such CGRAs. The compiler presented in this thesis is capable of accepting applications specified in C89 standard. To enable architectural design space exploration, the compiler is designed such that it can be customized for several instances of CGRAs employing dataflow execution paradigm at the orchestrator. This can be achieved by specifying the appropriate configuration parameters to the compiler. The focus of this thesis is to provide efficient support for various kinds of parallelism while ensuring correctness. The compiler is designed to support fine-grained task level parallelism that exists across iterations of loops and function calls. Additionally, compiler can also support pipeline parallelism, where a loop is split into multiple stages that execute in a pipelined manner. The prototype compiler, which targets multiple instances of a CGRA, is demonstrated in this thesis. We used this compiler to target multiple variants of CGRAs employing dataflow execution paradigm. We varied the reconfigur-able fabric, orchestration mechanism employed, size of instruction buffers. We also choose applications from two different domains viz. cryptography and linear algebra. The execution time of the CGRA (the best among all instances) is compared against an Intel Quad core processor. Cryptography applications show a performance improvement ranging from more than one order of magnitude to close to two orders of magnitude. These applications have large amounts of ILP and our compiler could successfully expose the ILP available in these applications. Further, the domain customization also played an important role in achieving good performance. We employed two custom functional units for accelerating Cryptography applications and compiler could efficiently use them. In linear algebra kernels we observe multiple iterations of the loop executing in parallel, effectively exploiting loop-level parallelism at runtime. Inspite of this we notice close to an order of magnitude performance degradation. The reason for this degradation can be attributed to the use of non-pipelined floating point units, and the delays involved in accessing memory. Pipeline parallelism was demonstrated using this compiler for FFT and QR factorization. Thus, the compiler is capable of efficiently supporting different kinds of parallelism and can support complete C89 standard. Further, the compiler can also support different instances of CGRAs employing dataflow execution paradigm.

This thesis is about the design of the telescopic platform. which should be fasten on robot called Breach. Purpose of this platform is to carry another equipment like sensors, cameras or manipulator. The platform has to provide chiefly the high demands on load capacity. Another part of the thesis is design of mechanism placed on this telescopic platform, or just on the robot Breach. This mechanism has to load up small subject on the top part of the platform and then unload it on chosen place. Only thing defined for this subjekt is maximal weight of 2 kilograms, but other parameters aren´t specified. Construction is therefore focused on maximal versatility of this mechanism.

Ces travaux de thèse visent à diminuer le coût de développement des applications mobilespour smartphones Android, iOS, etc. Les applications mobiles sont de plus en plus complexes. Auparavant, une application mobile se contentait d’afficher des données provenant du web. Maintenant, en plus de cela, elles communiquent avec le monde extérieur. Par exemple, certaines applications communiquent avec des montres, avec des écrans de télévision etc. D’autres permettent le scan de codes barres ou encore l’interaction avec des objets réels à travers la réalité augmentée. Les serveurs peuvent envoyer des notifications aux applications, etc. Une application mobile est devenue un logiciel à part entière. Cependant, pour toucher un maximum d’utilisateurs de smartphones, les applications mobiles doivent être conçues, implémentées et déployées sur tous les smartphones possibles. Avec la multiplication des configurations matérielles différentes ainsi que la multiplication des systèmes d’exploitation mobiles, cette tâche devient de plus en plus ardue. En effet, une application mobile doit souvent être réalisée une fois pour chaque plate-forme cible (Android, iOS, Windows Phone 8, etc.). Le temps et le coût de réalisation d’applications mobiles est donc multiplié par le nombre de plates-formes ciblées. Dans ces travaux, nous proposons de combiner le développement natif avec la programmation par composants. Pour ce faire, nous introduisons la notion de composants multiplateformes. Ce sont des composants qui peuvent être exécutés sur plusieurs plates-formes mobiles. Pour la représentation des composants, nous avons introduit la notion d’interface indépendante à n’importe quelle plate-forme mobile. Ainsi, l’intégration et l’assemblage se font d’une façon unique, que l’on soit dans un environnement de développement Android, iOS ou autre. Pour ce faire, nous avons spécifié un nouveau langage de programmation basé sur les annotations. Cette approche a été validée à travers le développement d’une application mobile pour Android et iOS avec notre solution. L’application implémentée a été réalisée en concordance avec les problématiques que rencontrent les entreprises de développement mobile et plus particulièrement Keyneosoft. Ensuite, nous avons comparé ces versions de l’application avec les versions développées nativement. Nous avons montré qu’avec notre solution nous diminuons le temps de développement d’au moins 30% sans aucune limitation pour les développeurs d’applications (même expérience utilisateur, même performance). Nous avons aussi comparé notre solution avec des produits disponibles sur le marché Phonegap, Titanium mobile et Xamarin. Nous en avons conclu que notre solutiton offrait le plus de possibilités sans aucune limitation. / In this thesis, we aim to decrease the development cost of applications for smartphones running Android, iOS, etc. Mobile applications are more and more complex. A few year ago, a mobile application was only design to display web content. Today, in addition, they are connected with the external world. For example, some applications are connected with watches, TVs, etc. Mobile applications became real softwares. However, in order to be visible by all smartphones users, mobile applications are designed, developed and deployed on every kind of smartphone. With the increase of the multiplicity of hardware configurations and the diversity of mobile operating systems, this task is becoming more and more laborious. Indeed, a mobile application is often implemented one time for each target platform (Android, iOS, Windows Phone 8, etc.). Therefore, the time and the cost for a mobile application implementation is multiplied by the number of target platforms. In this thesis, we propose to combine native development with the advantages of component-based software engineering. To do that, we have introduced the concept of multiplatform components. Those components are capable to be executed on any mobile platform. In order to describe components, we have introduced interfaces that are independent of any mobile platform. Thus, component integration and assembly are common on Android, iOS and others systems. To achieve that, we have specified a new programming language based on Annotations. We have validated this approach with the implementation of a real mobile application for Android and iOS. We have compared this application with the same application developed natively. Results show that with our solution, developers implement a multiplatform application 30% faster than native development. Moreover, our solution does not show any limitation for developers (same user experience, same performances). Finally, we have compared our solution with real products available on the software market: Phonegap, Titanium mobile and Xamarin. This comparison illustrates that our solution provides the best features and does not limit developers possibilities.