Refine
Year of publication
- 2017 (741) (remove)
Document Type
- Journal article (394)
- Doctoral Thesis (242)
- Complete part of issue (52)
- Preprint (12)
- Review (12)
- Book (8)
- Conference Proceeding (7)
- Jahresbericht (4)
- Book article / Book chapter (4)
- Master Thesis (2)
Language
- English (504)
- German (232)
- French (3)
- Multiple languages (2)
Keywords
- Würzburg (51)
- Wuerzburg (50)
- Wurzburg (50)
- Universität (48)
- University (45)
- Hadron-Hadron scattering (experiments) (28)
- High energy physics (26)
- medicine (14)
- physics (12)
- Medicine (11)
Institute
- Theodor-Boveri-Institut für Biowissenschaften (78)
- Universität - Fakultätsübergreifend (47)
- Physikalisches Institut (46)
- Graduate School of Life Sciences (35)
- Medizinische Klinik und Poliklinik I (25)
- Institut für Psychologie (24)
- Institut für Theoretische Physik und Astrophysik (24)
- Neurologische Klinik und Poliklinik (24)
- Medizinische Klinik und Poliklinik II (21)
- Fakultät für Physik und Astronomie (20)
Schriftenreihe
Sonstige beteiligte Institutionen
- Johns Hopkins School of Medicine, Baltimore, MD, U.S. (4)
- Institut für Optik und Atomare Physik, Technische Universität Berlin, 10623 Berlin, Germany (2)
- Laboratory for Chemistry and Life Science, Institute of Innovative Research, Tokyo Institute of Technology, Yokohama 226-8503, Japan (2)
- CERN (Geneva, Switzerland) (1)
- Deutsches Zentrum für Herzinsuffizienz (1)
- Didaktik der Chemie (1)
- Endokrinologie (1)
- Fraunhofer Institut für Silicatforschung ISC (1)
- Fraunhofer-Institut für Silicatforschung (1)
- Fraunhofer-Institut für Silicatforschung ISC (1)
ResearcherID
- D-3057-2014 (1)
- M-1240-2017 (1)
- N-8985-2015 (1)
Enterprise applications in virtualized data centers are often subject to time-varying workloads, i.e., the load intensity and request mix change over time, due to seasonal patterns and trends, or unpredictable bursts in user requests. Varying workloads result in frequently changing resource demands to the underlying hardware infrastructure. Virtualization technologies enable sharing and on-demand allocation of hardware resources between multiple applications. In this context, the resource allocations to virtualized applications should be continuously adapted in an elastic fashion, so that "at each point in time the available resources match the current demand as closely as possible" (Herbst el al., 2013). Autonomic approaches to resource management promise significant increases in resource efficiency while avoiding violations of performance and availability requirements during peak workloads.
Traditional approaches for autonomic resource management use threshold-based rules (e.g., Amazon EC2) that execute pre-defined reconfiguration actions when a metric reaches a certain threshold (e.g., high resource utilization or load imbalance). However, many business-critical applications are subject to Service-Level-Objectives defined on an application performance metric (e.g., response time or throughput). To determine thresholds so that the end-to-end application SLO is fulfilled poses a major challenge due to the complex relationship between the resource allocation to an application and the application performance. Furthermore, threshold-based approaches are inherently prone to an oscillating behavior resulting in unnecessary reconfigurations.
In order to overcome the deficiencies of threshold-based
approaches and enable a fully automated approach to dynamically control the resource allocations of virtualized applications, model-based approaches are required that can predict the impact of a reconfiguration on the application performance in advance. However, existing model-based approaches are severely limited in their learning capabilities. They either require complete performance models of the application as input, or use a pre-identified model structure and only learn certain model parameters from empirical data at run-time. The former requires high manual efforts and deep system knowledge to create the performance models. The latter does not provide the flexibility to capture the specifics of complex and heterogeneous system architectures.
This thesis presents a self-aware approach to the resource management in virtualized data centers. In this context, self-aware means that it automatically learns performance models of the application and the virtualized infrastructure and reasons based on these models to autonomically adapt the resource allocations in accordance with given application SLOs. Learning a performance model requires the extraction of the model structure representing the system architecture as well as the estimation of model parameters, such as resource demands. The estimation of resource demands is a key challenge as they cannot be observed directly in most systems.
The major scientific contributions of this thesis are:
- A reference architecture for online model learning in virtualized systems. Our reference architecture is based on a set of model extraction agents. Each agent focuses on specific tasks to automatically create and update model skeletons capturing its local knowledge of the system and collaborates with other agents to extract the structural parts of a global performance model of the system. We define different agent roles in the reference architecture and propose a model-based collaboration mechanism for the agents. The agents may be bundled within virtual appliances and may be tailored to include knowledge about the software stack deployed in a specific virtual appliance.
- An online method for the statistical estimation of resource demands. For a given request processed by an application, the resource time consumed for a specified resource within the system (e.g., CPU or I/O device), referred to as resource demand, is the total average time the resource is busy processing the request. A request could be any unit of work (e.g., web page request, database transaction, batch job) processed by the system. We provide a systematization of existing statistical approaches to resource demand estimation and conduct an extensive experimental comparison to evaluate the accuracy of these approaches. We propose a novel method to automatically select estimation approaches and demonstrate that it increases the robustness and accuracy of the estimated resource demands significantly.
- Model-based controllers for autonomic vertical scaling of virtualized applications. We design two controllers based on online model-based reasoning techniques in order to vertically scale applications at run-time in accordance with application SLOs. The controllers exploit the knowledge from the automatically extracted performance models when determining necessary reconfigurations. The first controller adds and removes virtual CPUs to an application depending on the current demand. It uses a layered performance model to also consider the physical resource contention when determining the required resources. The second controller adapts the resource allocations proactively to ensure the availability of the application during workload peaks and avoid reconfiguration during phases of high workload.
We demonstrate the applicability of our approach in current virtualized environments and show its effectiveness leading to significant increases in resource efficiency and improvements of the application performance and availability under time-varying workloads. The evaluation of our approach is based on two case studies representative of widely used enterprise applications in virtualized data centers. In our case studies, we were able to reduce the amount of required CPU resources by up to 23% and the number of reconfigurations by up to 95% compared to a rule-based approach while ensuring full compliance with application SLO. Furthermore, using workload forecasting techniques we were able to schedule expensive reconfigurations (e.g., changes to the memory size) during phases of load load and thus were able to reduce their impact on application availability by over 80% while significantly improving application performance compared to a reactive controller. The methods and techniques for resource demand estimation and vertical application scaling were developed and evaluated in close collaboration with VMware and Google.
This paper proposes an attitude determination system for small Unmanned Aerial Vehicles (UAV) with a weight limit of 5 kg and a small footprint of 0.5m x 0.5 m. The system is realized by coupling single-frequency Global Positioning System (GPS) code and carrier-phase measurements with the data acquired from a Micro-Electro-Mechanical System (MEMS) Inertial Measurement Unit (IMU) using consumer-grade Components-Off-The-Shelf (COTS) only. The sensor fusion is accomplished using two Extended Kalman Filters (EKF) that are coupled by exchanging information about the currently estimated baseline. With a baseline of 48 cm, the static heading accuracy of the proposed system is comparable to the one of a commercial single-frequency GPS heading system with an accuracy of approximately 0.25°/m. Flight testing shows that the proposed system is able to obtain a reliable and stable GPS heading estimation without an aiding magnetometer.
In the first part of his work, the causes for the sudden degradation of useable capacity of lithium-ion cells have been studied by means of complementary methods such as computed tomography, Post-Mortem studies and electrochemical analyses. The results obtained point unanimously to heterogeneous aging as a key-factor for the sudden degradation of cell capacity, which in turn is triggered by differences in local compression.
At high states of health, the capacity fade rate is moderate but some areas of the graphite electrode degrade faster than others. Still, the localized changes are hardly noticeable on cell level due to averaging effects. Lithium plating occurs first in unevenly compressed areas, creating patterns visible to the human eye. As lithium plating leads to rapid consumption of active lithium, a sudden drop in capacity is observed on cell level. Lithium plating appears to spread out from the initial areas over the whole graphite electrode, quickly consuming the remaining useful lithium and active graphite. It can be hypothesized that a self-amplifying circle of reciprocal acceleration of local lithium loss and material loss causes rapid local degradation.
Battery cell designers can improve cycle life by homogeneous pressure distribution in the cell and using negative active materials that are resilient to elevated discharge potentials such as improved carbons or lithium titanate. Also, a sufficiently oversized negative electrode and suitable electrolyte additives can help to avoid lithium plating. When packs are designed, care must be taken not to exert local pressure on parts of cells and to avoid both very high and low states of charge.
In the second part of this dissertation the resilience of cylindrical and pouchbag cells to shocks and different vibrations was investigated. Stresses inflicted by vibration and shock tests according to the widely recognized UN38.3 transport test were compared to a long-time test that exposed cells to a 186 days long ordeal of sine sweep vibrations with a profile based on real-world applications. All cells passed visual and electric inspection performed by TU München after the vibration tests. Only cylindrical cells subjected to long-term vibrations in axial direction showed an increase in impedance and a loss of capacity that could be recuperated in part.
The detailed analyses presented in this thesis gave more details on the damages inflicted by vibrations and shocks and revealed drastic damages in some cases. In cylindrical cells, only movement in axial direction caused damage. Long term vibrations were found to be especially detrimental.
No damage whatsoever could be detected for pouch cells, regardless of the test protocol and the direction of movement. The extreme resilience of pouchbag cells shows that the electrode stack of lithium-ion cells is resistant to vibrations, and that damages are caused by design imperfections that can be improved at low cost.
The findings of this work, and the general state of research show that it is most crucial to control the lithiation and thus potential of the graphite electrode.
In the last part of this work, a new, direct method for charge estimation based on changing transmission is presented. A correlation between transmission of short ultrasonic pulses and state of charge is found. This new technology allows direct measurement of the state of charge. The method is demonstrated for batteries with different positive active materials, showing its versatility. As the observed changes can be traced to the lithiation of graphite, it can be determined without a reference electrode. Already at this early stage of development, the found correlations allow estimation of state of charge. The present hysteresis in the signal height of the slow wave, which is unneglectable especially during discharging at higher currents, will be subject to further investigation.
The observed effects can be explained by effects on different length scales. Biot’s theory explains the second wave’s slowness based on the active material particles size in the range of 0.01 mm and electrolyte-filled pores. Lithiation of graphite changes the porosity of the electrode and thereby the velocity and wavelength of the impulse. When the wavelength approaches the length scale of the layers, 0.1 mm, scattering effects dampen the transmitted signal. Finally, the wavelength of the pulse should be shorter than the transducers diameter to obtain a homogeneous wave front.
To conclude, the new method allows the control of each individual cell in a pack independent from the electrical connections of the cells.
As the method shows great promise, further studies regarding factors such as long-term behavior, temperature and current rates should be conducted. In this thesis hysteresis was observed and a deeper understanding of the reasons behind it may allow further improvements of measurement precision.
Acute graft-versus-host disease (aGvHD) is a major cause of morbidity and mortality after allogeneic hematopoietic stem cell plus T cell transplantation (allo-HSCT). In this study, we investigated the requirement for CD28 co-stimulation of donor CD4\(^{+}\) conventional (CD4\(^{+}\)CD25\(^{-}\)Foxp3\(^{-}\), Tconv) and regulatory (CD4\(^{+}\)CD25\(^{+}\)Foxp3\(^{+}\), Treg) T cells in aGvHD using tamoxifen-inducible CD28 knockout (iCD28KO) or wild-type (wt) littermates as donors of CD4\(^{+}\) Tconv and Treg. In the highly inflammatory C57BL/6 into BALB/c allo-HSCT transplantation model, CD28 depletion on donor CD4\(^{+}\) Tconv reduced clinical signs of aGvHD, but did not significantly prolong survival of the recipient mice. Selective depletion of CD28 on donor Treg did not abrogate protection of recipient mice from aGvHD until about day 20 after allo-HSCT. Later, however, the pool of CD28-depleted Treg drastically declined as compared to wt Treg. Consequently, only wt, but not CD28-deficient, Treg were able to continuously suppress aGvHD and induce long-term survival of the recipient mice. To our knowledge, this is the first study that specifically evaluates the impact of CD28 expression on donor Treg in aGvHD. Moreover, the delayed kinetics of aGvHD lethality after transplantation of iCD28KO Treg provides a novel animal model for similar disease courses found in patients after allo-HSCT.
Einleitung: Die steigende Prävalenz adipöser Menschen führt weltweit zu einer relevanten Morbidität, die auch junge Frauen im geschlechtsreifen Alter betrifft. Damit gerät der Themenkomplex Adipositas und assoziierte Komplikationen auch im Hinblick auf die Versorgung Schwangerer in den Fokus. Das Ziel dieser Arbeit war es deshalb, die Adipositasprävalenz und hiermit assoziierte maternale und fetale Risikofaktoren zwischen 2006 und 2011 in einem lokalen Kollektiv zu untersuchen.
Material und Methoden: Die retrospektive Analyse umfasste alle maternalen und fetalen Daten von Patientinnen, die 2006 und 2011 an der Universitätsfrauenklinik Würzburg von einem Einling entbunden wurden. Die deskriptive Statistik umfasste die Prävalenz von Adipositas und Gewichtszunahme, maternale Risikofaktoren, Schwangerschaftskomplikationen und fetales Outcome.
Ergebnisse: Unsere Analyse umfasste 2838 Patientinnen mit Einlingsgraviditäten, die in den Jahren 2006 (n=1292) und 2011 (n=1545) an der Uniklinik Würzburg entbunden haben. Es zeigte sich, dass weder der initiale BMI noch die Gewichtszunahme während der Schwangerschaft zwischen 2006 und 2011 signifikant anstiegen. Die Mehrheit der übergewichtigen (71%) oder adipösen (60,4%) Patientinnen überstieg die empfohlene Gewichtszunahme. Die Prävalenz von adipositasassoziierten Erkrankungen wie Gestationsdiabetes und Präeklampsie stiegen signifikant an und waren mit einem hohen initialen BMI assoziiert. Während Übergewichtigkeit nicht mit einer Terminüberschreitung assoziiert war, wurden adipöse Patientinnen signifikant häufiger per Sectio caesarea entbunden. Das Geburtsgewicht war 2011 signifikant höher als 2006, wobei keine signifikanten Änderungen im fetalen Outcome dargestellt werden konnten.
Schlussfolgerung: Es gibt einen Trend zu vermehrter Gewichtszunahme während der Schwangerschaft. Assoziierte Risikofaktoren wie Gestationsdiabetes und Präeklampsie sind erhöht.
In transparent orthographies, persistent reading fluency difficulties are a major cause of poor reading skills in primary school. The purpose of the present study was to investigate effects of a syllable-based reading intervention on word reading fluency and reading comprehension among German-speaking poor readers in Grade 4. The 16-session intervention was based on analyzing the syllabic structure of words to strengthen the mental representations of syllables and words that consist of these syllables. The training materials were designed using the 500 most frequent syllables typically read by fourth graders. The 75 poor readers were randomly allocated to the treatment or the control group. Results indicate a significant and strong effect on the fluency of recognizing single words, whereas text-level reading comprehension was not significantly improved by the training. The specific treatment effect provides evidence that a short syllable-based approach works even in older poor readers at the end of primary school.
Visual saliency maps reflecting locations that stand out from the background in terms of their low-level physical features have proven to be very useful for empirical research on attentional exploration and reliably predict gaze behavior. In the present study we tested these predictions for socially relevant stimuli occurring in naturalistic scenes using eye tracking. We hypothesized that social features (i.e. human faces or bodies) would be processed preferentially over non-social features (i.e. objects, animals) regardless of their low-level saliency. To challenge this notion, we included three tasks that deliberately addressed non-social attributes. In agreement with our hypothesis, social information, especially heads, was preferentially attended compared to highly salient image regions across all tasks. Social information was never required to solve a task but was regarded nevertheless. More so, after completing the task requirements, viewing behavior reverted back to that of free-viewing with heavy prioritization of social features. Additionally, initial eye movements reflecting potentially automatic shifts of attention, were predominantly directed towards heads irrespective of top-down task demands. On these grounds, we suggest that social stimuli may provide exclusive access to the priority map, enabling social attention to override reflexive and controlled attentional processes. Furthermore, our results challenge the generalizability of saliency-based attention models.
Our ability of screening broad communities for clinically asymptomatic diseases critically drives population health. Sensory chewing gums are presented targeting the tongue as 24/7 detector allowing diagnosis by “anyone, anywhere, anytime”. The chewing gum contains peptide sensors consisting of a protease cleavable linker in between a bitter substance and a microparticle. Matrix metalloproteinases in the oral cavity, as upregulated in peri-implant disease, specifically target the protease cleavable linker while chewing the gum, thereby generating bitterness for detection by the tongue. The peptide sensors prove significant success in discriminating saliva collected from patients with peri-implant disease versus clinically asymptomatic volunteers. Superior outcome is demonstrated over commercially available protease-based tests in saliva. “Anyone, anywhere, anytime” diagnostics are within reach for oral inflammation. Expanding this platform technology to other diseases in the future features this diagnostic as a massive screening tool potentially maximizing impact on population health.
Background
While hypercholesterolemia plays a causative role for the development of ischemic stroke in large vessels, its significance for cerebral small vessel disease (CSVD) remains unclear. We thus aimed to understand the detailed relationship between hypercholesterolemia and CSVD using the well described Ldlr\(^{−/-}\) mouse model.
Methods
We used Ldlr\(^{−/-}\) mice (n = 16) and wild-type (WT) mice (n = 15) at the age of 6 and 12 months. Ldlr\(^{−/-}\) mice develop high plasma cholesterol levels following a high fat diet. We analyzed cerebral capillaries and arterioles for intravascular erythrocyte accumulations, thrombotic vessel occlusions, blood-brain barrier (BBB) dysfunction and microbleeds.
Results
We found a significant increase in the number of erythrocyte stases in 6 months old Ldlr\(^{−/-}\) mice compared to all other groups (P < 0.05). Ldlr\(^{−/-}\) animals aged 12 months showed the highest number of thrombotic occlusions while in WT animals hardly any occlusions could be observed (P < 0.001). Compared to WT mice, Ldlr\(^{−/-}\) mice did not display significant gray matter BBB breakdown. Microhemorrhages were observed in one Ldlr\(^{−/-}\) mouse that was 6 months old. Results did not differ when considering subcortical and cortical regions.
Conclusions
In Ldlr\(^{−/-}\) mice, hypercholesterolemia is related to a thrombotic CSVD phenotype, which is different from hypertension-related CSVD that associates with a hemorrhagic CSVD phenotype. Our data demonstrate a relationship between hypercholesterolemia and the development of CSVD. Ldlr\(^{−/-}\) mice appear to be an adequate animal model for research into CSVD.
Background:
The use of venoarterial extracorporeal membrane oxygenation (va-ECMO) via peripheral cannulation for septic shock is limited by blood flow and increased afterload for the left ventricle.
Case Report:
A 15-year-old girl with acute myelogenous leukemia, suffering from severe septic and cardiogenic shock, was treated by venoarterial extracorporeal membrane oxygenation (va-ECMO). Sufficient extracorporeal blood flow matching the required oxygen demand could only be achieved by peripheral cannulation of both femoral arteries. Venous drainage was performed with a bicaval cannula inserted via the left V. femoralis. To accomplish left ventricular unloading, an additional drainage cannula was placed in the left atrium via percutaneous atrioseptostomy (va-va-ECMO). Cardiac function recovered and the girl was weaned from the ECMO on day 6. Successful allogenic stem cell transplantation took place 2 months later.
Conclusions:
In patients with vasoplegic septic shock and impaired cardiac contractility, double peripheral venoarterial extracorporeal membrane oxygenation (va-va-ECMO) with transseptal left atrial venting can by a lifesaving option.