Concurrent data structures have found increasingly widespread use in both multicore and distributed computing environments, thereby escalating the priority for verifying their correctness. The thread safe behavior of these concurrent objects is often described using formal semantics known as linearizability, which requires that  every operation in a concurrent object appears to take effect between its invocation and response. Quasi linearizability is a quantitative relaxation of linearizability to allow more implementation freedom for performance optimization.  However, ensuring the quantitative aspects of this new correctness condition is an arduous task. We propose the first method for formally verifying quasi linearizability of the implementation model of a concurrent data structure. The method is based on checking the refinement relation between the implementation model and a specification model via explicit state model checking. It can directly handle multi-threaded programs where each thread can make infinitely many method calls, without requiring the user to manually annotate for the linearization points. We have implemented and evaluated our method in the PAT model checking toolkit.  Our experiments show that the method is effective in verifying quasi linearizability and in detecting its violations. / Master of Science

Indiana University-Purdue University Indianapolis (IUPUI) / Performance models are useful as mathematical models to reason about the behavior of different computer systems while running various applications. In this thesis, we aim to provide two distinct performance models: one for distributed-memory high performance computing systems with network communication, and one for deep neural networks. Our main goal for the first model is insight and simplicity, while for the second we aim for accuracy in prediction. The first model is generalized for networked multi-core computer systems, while the second is specific to deep neural networks on a shared-memory system.

The object of this diploma thesis is surveying and creation of 3D model of the church of st. Peter of Alcantara in Karviná. In the introduction this thesis is focused on history and localization of the church. In the next part is a description of data collection, computational work and graphic processing. The final results of the thesis are the wired model, which is created in MicroStation PowerDraft V8i, and the textured 3D model made in AutoCAD 2018.

El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado. / No presente resumen

Platelets are small and anucleate blood cells that are well known for their role in hemostasis. Other less recognized platelet functions include contributions to cancer vascularization, growth, and metastasis. Although the participation of platelets in cancer biology has been broadly studied in mouse models, there is no information in the literature regarding the crosstalk of platelets and cancer cells in dogs. The objective of this dissertation was to explore the interaction of canine platelets and tumor cells in vitro. The main hypothesis was that canine platelets were similar to human and murine platelets, and would interact with tumor cells, resulting in a change in the behavior of these cells. Using confocal immunofluorescence, we show that fibrinogen and von Willebrand factor have little colocalization within platelets, providing evidence that canine platelets have selective packaging and different alpha-granule subtypes, as shown in mice and humans. Then, we demonstrate canine platelet activation by osteosarcoma and mammary carcinoma cells, utilizing platelet aggregometry. Next, we show that intact platelets, platelet lysate, and thrombin-activated platelet releasate have an inhibitory effect on the migration of canine osteosarcoma and mammary carcinoma cell lines, contrary to what is described in humans and mice. We also demonstrate that releasate from canine platelets activated by collagen induces cancer cell migration, the opposite of the effect of releasates derived from thrombin or adenosine diphosphate activation. Lastly, we show that platelets can downregulate epithelial-to-mesenchymal-related transcription factors in canine cancer cells, suggesting that platelets may play an important role regulating this process in canine cancer progression. In conclusion, the results of this study have revealed important interactions between canine cancer cell lines and platelets in vitro. Our findings suggest that platelets most likely have a significant role in cancer growth in dogs and that cancer cells most likely affect platelet function in cancer patients.

Finite state machines are often used for modeling the decision logic for simulated systems. MathWorks’ Stateflow has a graphical user interface that allow users to model finite state machines. A Stateflow model can be added as a block to a Matlab/Simulink model and be executed seamlessly together. Stateflow blocks are developed as “charts” but they are natively stored as XML documents. This research explores the possibility of extracting the behavior of the finite state machines as defined in a Stateflow chart. This is done by parsing the corresponding XML document and reproducing this behavior in a C++ implementation that can be instantiated within a large, C++ based simulation system. Furthermore, the goal of this research is to develop a tool that will automatically generate an equivalent C++ representation, given an arbitrary Stateflow XML model. This research is performed in the context of developing highidelity powertrain simulations to be executed in High-Performance Computing environments.

In this paper, it is clearly shown that the distribution of the initial porosity is a critical factor in the prediction of damage evolution and initiation of failure in a cast AM60B magnesium notch Bridgeman tensile specimen. Using X-ray computed tomography, the actual initial porosity distribution was obtained, and this distribution was input into a finite element code as an initial condition. The predicted damage evolution from this simulation was compared to the damage evolution of the experimental specimen as well as other simulated porosity distributions. This study shows that the simulation of the actual porosity distribution predicted well the damage evolution observed in the experiment. It is also shown that the initial distribution of porosity plays a vital role in the predicted elongation to failure of a notched specimen. The actual distribution was shown to fail at a significantly lower strain than random or uniformly distributed damage.

Emergency managers are faced with the challenge of acting quickly after a hurricane but rarely have detailed information available about type and amount of damage. In response to this need, linear additive geospatial models based on logistic regression analyses of driving variables including wind, rain, surge, topography were developed and automation routines programmed that rapidly and accurately predict a variety of damage types. Since a preponderance of damage is associated with falling trees, over 2000 post-Katrina forested plots were used to fit and validate independent models for hardwood blowdown and pine shear. Additional models using peak wind gusts and maximum sustained winds respectively were fully automated. Most importantly, total model run time was decreased from 36 to 5 hours for the more complicated forest damage models. The models have been vetted by the Mississippi Emergency Management Agency (MEMA) and will be part of MEMA’s hurricane action response plans.

No description available.

PhD Thesis (Software Engineering) / Model-Based Development (MBD) is an approach that uses software models to describe the behaviour of embedded software and cyber-physical systems. MBD has become an increasingly prevalent paradigm, with Simulink by MathWorks being the most widely used MBD platform for control software. Unlike textual programming languages, visual languages for MBD such as Simulink use block diagrams as their syntax. Thus, some software engineering principles created for textual languages are not easily adapted to this graphical notation or have not yet been supported. A software engineering principle that is not readily supported in Simulink is the modularization of systems using information hiding. As with all software artifacts, Simulink models must be constantly maintained and are subject to evolution over their lifetime. This principle hides likely changes, thus enabling the design to be robust with respect to future changes. In this thesis, we perform repository mining on an industry change management system of Simulink models to understand how they are likely to change. Then, we explore the various modelling mechanisms available in the Simulink language to see how they could support modular design with information hiding. Next, we propose a module structure, syntactic interface, and modelling conventions for Simulink designs, which are supported by our open-source Simulink Module Tool. Finally, we apply the proposed techniques on case studies from the aerospace and nuclear domains, in order to demonstrate their practicality and validate their effectiveness. Overall, the approach helped support information hiding by encapsulating secrets and facilitating likely changes. It also had a positive effect on interface complexity, cohesion, and coupling. The larger system also exhibited reductions to cyclomatic complexity, testing effort, and execution time, but the smaller case study benefited less in these areas. / Thesis / Doctor of Philosophy (PhD)