Commodity supercomputing clusters known as Beowulf clusters, have become a low cost alternative to traditional supercomputers. Beowulf clusters combine inexpensive computers and specialized software to achieve supercomputing power. The processing nodes in a diskless Beowulf cluster do not have a local hard disk unlike the nodes in most commodity clusters. Research has provided performance information for diskless clusters built with expensive, high performance equipment. Beowulf clusters use commodity off-the-shell hardware, and little information is available about their performance. This research includes the construction of several diskless Beowulf clusters. Using the NAS Parallel Benchmarks, the performance of these clusters was measured. Through analysis of these measurements, a descriptive performance model of diskless Beowulf clusters was produced.

In the research regarding rating ESL writing assessments, borderline writing samples are mentioned, but a solution has yet to be addressed. Borderline samples are writing samples that do not perfectly fit a set level within the rubric, but instead have characteristics from multiple levels. The aim of this thesis is to provide an improved training module in the setting of an Intensive English Program by exposing new raters to borderline samples and rating rationale from experienced raters. The purpose of this training is to increase the confidence, consistency, and accuracy of novice raters when rating borderline samples of writing. The training consists of a workbook with a rubric and instructions for use, benchmark examples of writing, borderline examples of writing with comments from experienced raters defending the established scores, then a variety of writing samples for practice. The selection of the benchmark and the borderline examples of writing was informed by the fit statistic from existing datasets that had been analyzed with many-facet Rasch measurement. Eight experienced raters participated in providing rationale based upon the rubric explaining why each borderline sample was rated with its established score, and describing why the sample could be considered at a different level. In order to assess the effectiveness of the training workbook, it was piloted by 10 novice raters who rated a series of essays and responded to a survey. Results of the survey demonstrated that rater confidence increased following the training, but that they needed more time with the training materials to use them properly. The statistical analyses showed insignificant changes, which could be due to the limitations of the data collection. Further research regarding the effectiveness of this training workbook is necessary, as well as an increased discussion in the field regarding the prevalent issue of rating borderline samples of writing.

The summer after Amanda S. Frasier left K-12 teaching to return to high education, she received an email stating that her students’ end-of-year assessments were in. When she looked at the scores, she was pleased to see they had done so well, even though she questions the value of standardized accountability measures and didn’t learn anything new from the results. Frasier discusses why these scores are not helpful and advocates more meaningful measures.

Kubernetes is a container orchestration system that has been rising in popularity in recent years. The modular nature of Kubernetes allows the usage of different storage solutions, and for cloud environments, cloud-native distributed storage solutions maybe attractive due to their redundant nature. There are many tools for cloud-native distributed storage available on the market today with differing features and performance. Choosing the correct one for an organisation can be difficult. Organisations utilising Kubernetes in cloud environments would like to be as performance efficient as possible to save on costs and resources. This study aims to offer a benchmark and analysis for some of the most popular tools, to help organisations choose the ‘best’ solution for their operational needs, from a performance perspective. The benchmarks compare three cloud-native distributed storage solutions, OpenEBS, Portworx, and Rook-Ceph on both Amazon Elastic Kubernetes Service (EKS) and Azure Kubernetes Service (AKS). For a baseline comparison, the study will also benchmark the cloud providers own solutions; Azure Disk Storage, and Amazon Elastic Block Storage. The study compares these solutions from three key metrics; bandwidth, latency, and IOPS, in both read and write performance. / <p>Det finns övrigt digitalt material (t.ex. film-, bild- eller ljudfiler) eller modeller/artefakter tillhörande examensarbetet som ska skickas till arkivet.</p><p>There are other digital material (eg film, image or audio files) or models/artifacts that belongs to the thesis and need to be archived.</p>

Syftet med detta arbete är att undersöka hur överavkastning på aktiebörsen kan uppnås, genom att applicera investeringsstrategier som utgår ifrån portföljanalys. Studien inriktar sig på fastighetsbolag där svenska, europeiska och amerikanska bolag ingår. Metoden utgår ifrån att vikta innehaven i portföljerna genom användningen av finansiella nyckeltal. De beaktade nyckeltalen utgörs av P/E, EV/S, ROIC och soliditet. Portföljerna viktas periodvis utefter ett antal hypoteser som tagits fram. Hypoteserna berör vilka värden på respektive nyckeltal som anses vara mest fördelaktiga ur ett investeringsperspektiv. Studien syftar till att ta fram vilka nyckeltal som är kritiska för en överavkastning i fastighetsaktier. Utfallet av undersökningen visade på en överavkastning i åtta av elva portföljer, vilket togs fram genom användandet av jämförelseindex. Studien ledde fram till slutsatsen att en av fyra hypoteser inte kunde förkastas, medan övriga förkastades med bakgrund av det kvantitativa resultatet. Genom tillämpning av den kvantitativa metod som genomförs i vår studie påvisas att EV/S är det enda nyckeltalet, av urvalet, som en investeringsstrategi kan baseras på för att uppnå en överavkastning gentemot ett relevant index. / This study revolves around reaching high returns on the stock exchange by applying investment strategies based on portfolio analysis. The research focuses on real estate companies within the Swedish, European and American stock exchange. By using financial ratios, P/E, EV/S, ROIC and Equity Ratio, the portfolios and its returns are optimized by a method of weighing. The shares within the portfolios are weighed periodically depending on a variety of hypotheses regarding the financial ratios and its values. The hypotheses have been developed to maximize the returns from an investment perspective. The study is trying to reach a consensus of which financial ratios are crucial within real estate companies to reach a high return. By using different indexes, the result showed that eight out of eleven portfolios reached an excess return. A conclusion was reached; one of four financial ratios are of great importance when trying to reach an excess return by investing in real estate companies. By applying the study’s quantitative method, it is shown that EV/S is the only financial ratio out of the selected that can be used in this investment strategy to achieve excess return, when compared to relevant indexes.

No description available.

Single-cell RNA-sequencing makes possible to study the gene expression at the level of individual cells. However, one of the main challenges of the single-cell RNA-sequencing analysis today, is the identification and annotation of cell types. The current method consists in manually checking the expression of genes using top differentially expressed genes and comparing them with related cell-type markers available in scientific publications. It is therefore time-consuming and labour intensive. Nevertheless, in the last two years,numerous automatic cell-type identification and annotation tools which use different strategies have been created. But, the lack of specific comparisons of those tools in the literature and especially for immuno-oncologic and oncologic purposes makes difficult for laboratories and companies to know objectively what are the best tools for annotating cell types. In this project, a review of the current tools and an evaluation of R tools were carried out.The annotation performance, the computation time and the ease of use were assessed. After this preliminary results, the best selected R tools seem to be ClustifyR (fast and rather precise) and SingleR (precise) for the correlation-based tools, and SingleCellNet (precise and rather fast) and scPred (precise but a lot of cell types remains unassigned) for the supervised classificationtools. Finally, for the marker-based tools, MAESTRO and SCINA are rather robust if they are provided with high quality markers.

The private sector has an increasingly important role in limiting the temperature rise to below the crucial 1.5°C of the Paris Agreement. In order to assess how well a portfolio is aligned to the goals of the Paris Agreement, benchmarks need be used. The creation of benchmarks is complex and varying methodologies with associated opinions exist. The purpose of this thesis is to examine two different approaches for creating benchmarks; Single scenario and Warming function. The usability of the benchmarks is examined from the perspective of a financial institution. The methodology is based on a literature review of earlier research. Using empirical data the different methods are tested in a smaller Case study and on a larger data set. The results give insights in different advantages and disadvantages of the methods. Using a Single scenario approach allows for a more granular analysis, however it is criticized for being misleading as the result is easily manipulated. The Warming function is less granular, as almost all companies in the empirical data are assessed against the same benchmark - creating unfair results for companies active in sectors with different demands on emission reduction. In order to gain sectoral granularity the Warming function methodology need to be investigated further. By lowering the demands on the final benchmarks, the adaptation to different sectors can increase inthe Warming function approach. / Den privata sektorn har en växande roll i kampen för att begränsa den globala uppvärmningen till Parisavtalets kritiska gräns på 1,5°C. För att utvärdera hur väl ett företag presterar klimatmässigt jämfört med kraven i Parisavtalet behövs jämförelsemått. Skapandet av jämförelsemått är komplext och en mängd olika metodologier existerar med olika grader av popularitet. Syftet med denna uppsats är att undersöka två olika tillvägagångssätt för att skapa jämförelsemått; Single scenario och Warming function. Användarbarheten av metodologierna utvärderas från perspektivet av en finansiell institution. Metodologin baseras på en litteraturstudie av tidigare undersökningar. Warming function och Single scenario utvärderas i en mindre fallstudie och i tester på ett större data set med hjälp av empiriska data. Resultatet ger insikter i olika för- och nackdelar med metoderna. Att använda Single scenario tillåter en mer granulär analys, dock kritiseras denna metod för att vara missledande då de slutgiltiga resultaten är enkla att manipulera. Warming functio när mindre granulär då de flesta företagen utvärderas mot samma jämförelsemått vilket skapar orättvisa resultat för företag som är aktiva i sektorer med olika höga krav på utsläppsminskning. För att öka granulariteten måste metodologin för Warming function undersökas djupare. Genom att sänka kraven på de slutgiltiga jämförelsemåtten kan sektoranpassade jämförelsemått öka i Warming function.

Portfolio allocation theories have been studied and used ever since the mid 20th century. Nevertheless, many investors still rely on personal expertise and information gathered from the market when building their investment portfolios. The purpose of this master’s thesis is to examine how personal preferences and expertise perform compared to mathematical portfolio alloca- tion theories and how the risk between these di↵erent strategies di↵er. Using two portfolio allocation theories, the Black-Litterman model and mod- ern portfolio theory (Markowitz), a portfolio managed by the investment firm Placerum Kapitalf ̈orvaltning in Ume ̊a will be compared and challenged to investigate which strategy gives the best risk adjusted return. Using scenario modelling, the portfolios can be compared using both historical data and future forecasted scenarios to analyze the past, present and future of the allocation theories and Placerum’s investment strategy. The first allocation theory, the Black-Litterman model, combines historical information from the market with views and preferences of the investor to select the optimal allocations derived from return and volatility. The second allocation theory, the modern portfolio theory (Markowitz), only uses histori- cal data to derive correlations and returns which are then used to select the optimal allocations. By analysing several risk measures applied on the portfolios historical and forecasted data as well as comparing the performance of the portfolios, it is shown that the investment strategy used at Placerum succeeds with its intentions to achieve relatively high return while reducing the risk. However, the portfolios given using the two allocation theories results in higher potential returns but at the cost of taking on a higher risk. Comparing the two studied allocation theories, it is shown that when using the Black-Litterman model with the assumptions and views defined in this project, modern allocation theory actually beats it in terms of potential return as well as in terms of risk adjusted return, even though its underlying theory is much simpler.

Insomni är ett vanligt problem och det finns behov av ökad tillgänglighet till kostnadseffektiva behandlingar. Syftet var att undersöka om en enkomponentsbehandling (EK), bestående av sömnrestriktion/sömnkomprimering, var non-inferior till en multikomponent (MK) KBT- behandling vid insomni och om det fanns en skillnad i symtomreduktion. Gränsvärden för non- inferiority var d = 0.8, utifrån tidigare forskning, respektive en strängare gräns d = 0.4. Data från en forskningsstudie där deltagarna erhöll EK (n = 193) jämfördes mot en riktlinje i form av en KBT-behandling i reguljärvården, MK (n = 289). Båda grupper erhöll behandling via samma internetplattform, och Insomnia Severity Index (ISI) användes som utfallsmått. Resultaten visade att EK inte kunde bekräftas som non-inferior till MK direkt efter behandling eller vid 1-årsuppföljning när en sträng gräns användes. När en liberal gräns användes kunde EK bekräftas som non-inferior direkt efter behandling men inte ett år senare. Direkt efter behandling och vid 1-årsuppföljningen visade båda grupperna en signifikant minskning av insomnisymtom, men vid 1-årsuppföljningen visade MK en större minskning. Ett stort bortfall vid 1-årsuppföljning och mer terapeutstöd för MK kan ha påverkat resultaten. Fler RCT-studier med långtidsuppföljningar behövs inom området, och även studier på andra populationer. MK är att föredra, men vid begränsade resurser kan EK erbjudas med god effekt på både kort och lång sikt. / Insomnia is a common problem and there is a need for increased accessibility to cost-effective treatments. The purpose was to examine if an one-component treatment (EK), consisting of sleep-restriction/sleep-compression, were non-inferior to a multi-component (MK) CBT treatment for insomnia and if there was a difference in symptom reduction. The prestated margins for non-inferiority were d = 0.8, based on previous research, and a stricter margin of d = 0.4. Data from a research study where participants received EK (n = 193) was compared to a benchmark consisting of a CBT treatment in regular health care (MK) (n = 289). Both groups received treatment at the same Internet platform, and Insomnia Severity Index (ISI) was used as outcome measurement. The results showed that EK could not be confirmed as non-inferior to MK directly after treatment or at the one-year follow up when using a strict limit. When a liberal limit was used, EK could be confirmed as non-inferior directly after treatment but not one year later. Directly after treatment and at the one-year follow up both groups showed a significant symptom reduction, but at the one-year follow up MK showed a greater reduction. A large number of missing data at the one-year follow up and more therapeutic support in MK may have affected the results. More RCT-studies with longterm follow ups are needed in the area, and also studies of other populations. MK is preferable, but with limited resources EK can be offered with good effect both short term and long term.