To build a shared-memory programming model for FPGAs, a fast and highly parallel method of accessing the shared-memory is required. This thesis presents a first look at how to implement a coherent caching system in an FPGA. The coherent caching system consists of multiple distributed caches that implement the write-once coherence protocol, allowing efficient access to system memory while simplifying the user programming model. Several test applications are used to verify functionality, and assess performance of the current system. Results show that with a processor-based system, some applications could benefit from improvements to the coherence system, but for many applications, the current system is sufficient. However, the current coherent caching system is not sufficient for most hardware core based systems, because the faster memory accesses quickly saturate shared system resources. As well, the performance of distributed-memory systems currently surpasses that of the coherent caching system. Performance results are promising, and given the potential for improvements, future work on this system is warranted.

To build a shared-memory programming model for FPGAs, a fast and highly parallel method of accessing the shared-memory is required. This thesis presents a first look at how to implement a coherent caching system in an FPGA. The coherent caching system consists of multiple distributed caches that implement the write-once coherence protocol, allowing efficient access to system memory while simplifying the user programming model. Several test applications are used to verify functionality, and assess performance of the current system. Results show that with a processor-based system, some applications could benefit from improvements to the coherence system, but for many applications, the current system is sufficient. However, the current coherent caching system is not sufficient for most hardware core based systems, because the faster memory accesses quickly saturate shared system resources. As well, the performance of distributed-memory systems currently surpasses that of the coherent caching system. Performance results are promising, and given the potential for improvements, future work on this system is warranted.

We consider asynchronous multiprocessors where processes communicate only by reading or writing shared memory. We show how to implement consensus, all comparison primitives (such as CAS and TAS), and load-linked/store-conditional using only a constant number of remote memory references (RMRs), in both the cache-coherent and the distributed-shared-memory models of such multiprocessors. Our implementations are blocking, rather than wait-free: they ensure progress provided all processes that invoke the implemented primitive are live. Our results imply that any algorithm using read and write operations, comparison primitives and load-linked/store-conditional, can be simulated by an algorithm that uses read and write operations only, with at most a constant-factor increase in RMR complexity.

Researchers interested in examining a given species of interest (or target species) that lacks complete sequence data can infer some knowledge of that species from one or more related species that has a complete set of data. To infer knowledge, it is desired to compare the available sequence data between the two species to find orthologs. However, without complete data sets, one cannot be certain of the validity of the detected orthologs. Using ortholog detection systems in concert with species’ mapping data, researchers can find regions of shared synteny, allowing for more certainty of the detected orthologs as well as allowing inference of some genetic information based on these regions of shared synteny. A pipeline software solution, Detection of Orthologs via Genetic Mapping Augmentation (DOGMA), was developed for this purpose. DOGMA’s functionality was tested using a target species, Phaseolus vulgaris, which only had partial sequence data available, and a closely related species, Glycine max, which has a fully se- quenced genome. On sequence similarity alone, which is the standard technique for detecting or- thologs, 205 potential orthologs were detected. DOGMA then filtered these results using mapping data from each species to determine that 121 of the 205 were quite likely true orthologs, referred to as putative orthologs, and the remaining 84 were categorized as reduced orthologs as there was either insufficient information present or were clearly outside a noted region of shared synteny. This provides evidence that DOGMA is capable of reducing false positives versus traditional techniques, such as applications based on Reciprocal Best BLAST Hits. If we interpret the output of the Or- tholuge program as the correct answer, DOGMA achieves 95% sensitivity. However, it is possible that some of the reduced orthologs classified by DOGMA are actually Ortholuge’s false positives, since DOGMA is using mapping data. To support this idea, we show DOGMA’s ability to detect false positives in the results of Ortholuge by artificially creating a paralog and removing the real ortholog. DOGMA properly classifies this data as opposed to Ortholuge.

The functional connection between the building and the streetscape affects the city and its attractiveness. Stores located in shopping malls with their entrances facing internal walkways, creates segregation between the mall and the streetscape. This segregation implies that the shopping mall loses some of its attractiveness, which may result in a so-called Greyfield mall. The expression Greyfield mall is used for shopping malls that have lost its visitors. Competition from new modern malls and a lack of investments are examples of factors that influence the development of Greyfiled malls. To clarify the relationship between the building and the streetscape, Space syntax has been used. Space syntax can be explained as the order between the different spaces. A Space syntax analysis provides an intergration value that indicates how the streets are linked. The pupose of this paper is to investigate and develop methods that will increase the attractiveness of shopping centers and malls in danger of developing into, or already classified as Greyfield malls. Improvements to these sites contributes to a better urban environment. The question formulation has been focused on the streetscape- and building design, and the movement patterns of the visitors. The questions have been answered through literature studies within the areas of; Space syntax, Shared space and Greyfield malls. In addition, a new proposal for the design of the building and the surrounding streetscape has been done. Analyses using Depth map has been used in a case study of Rosengallerian. Rosengallerian is located in the center of the city Husqvarna and has potential to act as a gathering place for city residents. By using Shared space, Woonerf-streets and axial analyzes with Space syntax a new proposal for mall and its surroundings has been developed. The new proposal is based on the prioritation of the pedestrians instead of the motor traffic. The streets surrounding the shopping mall has been converted into walking-speed areas, where the streets have been given a more narrow character. These changes creates a more attractive environment for the pedestrians around the mall. The results of the studies shows that the problem mainly lies in the design of the building and the priorities of the surrounding street network. The planning and prioritations of a street network are important to city residents, when the traffic moves according to the terms of the pedestrians a center where people can move freely is created. The ability to move around freely along with the city's walkways being extended into the building means that people wants to come back to the shopping mall. The proposal increase the attractiveness of the shopping mall since people attracts other people. The conclusion is that the mall is to be seen as an individual building being fitted into the existing surrounding environment. The shopping mall should be a natural part of the city's movement pattern, but also allowed to stand out architecturally. / Den funktionella kontakten mellan byggnaden och gaturummet påverkar staden och dess attraktivitet. Butiker som är placerade i gallerior och som har sina entréer mot interna stråk skapar en segregation mellan gallerian och gaturummet. Denna segregering innebär att gallerian förlorar en del av sin attraktivitet, vilket kan resultera i en så kallad Greyfield mall. Uttrycket Greyfield mall används för gallerior som förlorat sina besökare. Konkurrensen från nya moderna gallerior samt brist på investeringar är exempel på faktorer som påverkar uppkomsten av Greyfield malls. För att klargöra sambandet mellan byggnaden och gaturumet har ”Space syntax” använts. Space syntax kan förklaras som ordningen mellan de olika rummen. En Space syntax-analys ger ett integrationsvärde som visar hur stadens gator hänger samman. Syftet med detta examensarbete är att undersöka och utveckla metoder för att öka attraktiviteten hos köpcentrum och gallerior som riskerar att utvecklas till, eller redan klassas som Greyfield malls. Förbättringar av dessa platser bidrar till en godare stadsmiljö. Frågeställningarna har inriktats på gatans och byggnadens utformning samt besökarnas rörelsemönster. Frågeställningarna har besvarats genom litteraturstudier inom områdena Space syntax, Shared space och Greyfield malls. Därtill har ett nytt förslag på utformning av gatunätet och byggnaden utförts. Analyser med hjälp av Depth map har gjorts i en fallstudie av Rosengallerian. Rosengallerian är placerad i Huskvarna centrum och har potential att fungera som en samlingspunkt för stadens invånare. Områdets nuvarande utformning gynnar biltrafiken vilket har resulterat i en nedprioritering av gångtrafikanterna. Med hjälp av; Shared space, Boone-gator och axialanalyser med Space syntax har ett nytt förslag för Rosengallerian och dess omgivning tagits fram. Förslaget bygger på att stadens gångtrafikanter istället prioriteras. Gatorna runt gallerian har omvandlats till gångfartsområde, där gatorna har fått en smalare karaktär. Dessa förändringar skapar en attraktivare miljö för gångtrafikanterna kring Rosengallerian. För att öppna upp gallerian har de ursprungliga gångarna inne i gallerian ändrats till två raka tvärgator igenom byggnaden som sammanbundits med den övriga staden. Resultatet av studierna visar att problemet huvudsakligen ligger i byggnadens utformning samt det omkringliggande gatunätets prioriteringar. Ett gatunäts planering och prioriteringar är av vikt för stadens invånare, då biltrafiken förflyttar sig på de gåendes villkor skapas ett centrum där människor kan röra sig fritt. Den fria rörelsen tillsammans med att stadens gator förlängs in i byggnaden medför att invånarna vill återkomma till gallerian. Förslagets förändringar ökar gallerians attraktivitet då människor attraherar andra människor. Slutsatsen är att gallerian skall ses som en enskild byggnad som skall passas in i den omkringliggande befintliga miljön. Gallerian bör vara en naturlig del av stadens rörelsemönster samtidigt som den gärna får markeras arkitektoniskt.

Bakgrund: Shared services är en strategi som innebär att en existerande verksamhetsfunktion, som tidigare varit spridd på flera olika platser, koncentreras till en enda enhet (center) inom en koncern. Det har blivit allt vanligare att koncerner väljer att upprätta ett SSC, avseende ekonomifunktionen, då det är ett sätt att effektivisera koncernens redovisningsprocesser. Till följd av denna utveckling har vissa revisorer påtalat problem kopplade till revisionen av företag inom en koncern som har ett SSC. Problem som kan uppkomma vid gränsöverskridande revisionsarbete är exempelvis bristande förståelse för lokala redovisningsregler och skatter samt språkskillnader som kan försvåra revisorns kontroll. Även problem rörande arbetsfördelning och arbetsinsatser kan behöva lösas. Trots att revisorer har påpekat denna typ av problem verkar det ha utförts få studier kring kopplingen mellan SSC och revision. Denna brist inom området gör det intressant att studera hur förekomsten av ett SSC påverkar revisionen. Syfte: Syftet med denna studie är att utreda om en koncerns SSC leder till några problem kopplade till revisionen av koncernens företag. Syftet behandlas både utifrån ett företags- och ett revisorsperspektiv. Metod: Denna studie är en tvärsnittsstudie mellan två grupper: företag inom koncerner med etablerade SSC:s samt personer som arbetar med revision av företag inom en koncern med ett SSC. Tre företagsrespondenter och fyra respondenter från revisionsbyråer har medverkat i studien genom intervjuer och skriftliga frågeformulär. Slutsats: Efter avslutad studie kan vi konstatera att ett SSC medför vissa problem kopplade till revisionen. Vi har kunnat urskilja flera typer av svårigheter, men framförallt har två övergripande problemområden framträtt: kommunikation samt arbets- och ansvarsfördelning. Dessa problemområden är starkt kopplade till varandra, då en god arbets- och ansvarsfördelning bygger på en god kommunikation. Kommunikationsproblem förekommer både revisorer emellan samt mellan revisorer och det granskade företaget. Då den ekonomiska informationen för en koncerns olika bolag finns samlad i ett SSC blir SSC:t ytterligare en part i revisionen att ta i beaktande för alla medverkande revisorer, såväl i moderbolaget som i dotterbolagen.

Regularity is a shared memory consistency condition that has received considerable attention, notably in connection with quorum-based shared memory. Lamport's original definition of regularity assumed a single-writer model, however, and is not well defined when each shared variable may have multiple writers. In this thesis, we address this need by formally extending the notion of regularity to a multi-writer model. We have shown that the extension is not trivial. While there exist various ways to extend the single-writer definition, the resulting definitions will have different strengths. Specifically, we give several possible definitions of regularity in the presence of multiple writers. We then present a quorum-based algorithm to implement each of the proposed definitions and prove them correct. We study the relationships between these definitions and a number of other well-known consistency conditions, and give a partial order describing the relative strengths of these consistency conditions. Finally, we provide a practical context for our results by studying the correctness of two well-known algorithms for mutual exclusion under each of our proposed consistency conditions.

Shared objects are a key component in today's large distributed systems. Linearizability is a popular consistency condition for such shared objects which gives the illusion of sequential execution of operations. The time bound of an operation is the worst-case time complexity from the operation invocation to its response. Some time bounds have been proved for certain operations on linearizable shared objects in partially synchronous systems but there are some gaps between time upper bound and lower bound for each operation. In this work, the goal is to narrow or eliminate the gaps and find optimally fast implementations. To reach this goal, we prove larger lower bounds and show smaller upper bounds (compared to 2d for all operations in previous folklore implementations) by proposing an implementation for a shared object with an arbitrary data type in distributed systems of n processes in which every message delay is bounded within [d-u, d] and the maximum skew between processes' clocks is epsilon. Considering any operation for which there exist two instances such that individually, each instance is legal but in sequence they are not, we prove a lower bound of d + min{epsilon, u, d/3}, improving from d, and show this bound is tight when epsilon < d/3 and epsilon < u. Considering any operation for which there exist k instances such that each instance separately is legal and any sequence of them is legal, but the state of the object is different after different sequences, we prove a lower bound of (1-1/k)u, improving from u/2, and show this bound is tight when k = n. A pure mutator only modifies the object but does not return anything about the object. A pure accessor does not modify the object. For a pure mutator OP1 and a pure accessor OP2, if given a set of instances of OP1, the state of the object reflects the order in which the instances occur and an instance of OP2 can detect whether an instance of OP1 occurs, we prove the sum of the time bound for OP1 and OP2 is at least d + min{epsilon, u, d/3}, improving from d. The upper bound is d + 2*epsilon from our implementation.

Nowadays, organizations are facing more competitive environment under the impacts of globalization and technology. How to respond, forecast and keep their competitive advantage becomes very important and urgent for them to figure out. This impacts HR as well. As Wright & Snell pointed out HR faces three challenges now. The first challenge is value creation. In order to create value in global economy, ¡§HR executive must understand clearly the confluence of economic, social, and technological forces that drive industry competition¡¨ (Wright & Snell, 2005). The second challenge is value delivery. It becomes more urgent and vital for HR to demonstrate its tangible impacts by aligning with and driving the issues critical to business to deal with the pressing need (Wright & Snell, 2005). The third challenge is living value for HR profession. HR profession has to rediscover that they are not just order takers or implementers, but are guardians of organizations. Therefore, how to create more added-value becomes a very popular topic among HR filed especially for share-serviced organizations nowadays. Shared services is the idea that integrating resources in one place to reach economic of scale and serve to all countries of one region. This dissertation takes a qualitative approach to explore how shared-service organizations create their HR value by interviewing five muti-national organizations. The main argument of this dissertation is to discuss what happened on HR structure and role of Taiwan subsidiary after outsourcing the transactional part to create more value to organizations. There are three objectives in this dissertation. The first objective is the impacts on structure of Taiwan HR. The second objective is the impacts on Taiwan HR role and value creation. The final objective is delivery channel. Through interviewing shared-service organization, it enable us to have more understanding how they create their HR value and the impacts on HR structure and their roles in organizations.

Regularity is a shared memory consistency condition that has received considerable attention, notably in connection with quorum-based shared memory. Lamport's original definition of regularity assumed a single-writer model, however, and is not well defined when each shared variable may have multiple writers. In this thesis, we address this need by formally extending the notion of regularity to a multi-writer model. We have shown that the extension is not trivial. While there exist various ways to extend the single-writer definition, the resulting definitions will have different strengths. Specifically, we give several possible definitions of regularity in the presence of multiple writers. We then present a quorum-based algorithm to implement each of the proposed definitions and prove them correct. We study the relationships between these definitions and a number of other well-known consistency conditions, and give a partial order describing the relative strengths of these consistency conditions. Finally, we provide a practical context for our results by studying the correctness of two well-known algorithms for mutual exclusion under each of our proposed consistency conditions.