004 Datenverarbeitung; Informatik
Refine
Has Fulltext
- yes (571) (remove)
Year of publication
Document Type
- Article (176)
- Monograph/Edited Volume (162)
- Doctoral Thesis (131)
- Postprint (48)
- Conference Proceeding (38)
- Master's Thesis (10)
- Preprint (3)
- Bachelor Thesis (1)
- Habilitation Thesis (1)
- Moving Images (1)
Language
- English (407)
- German (162)
- Multiple languages (2)
Keywords
- Informatik (21)
- Didaktik (15)
- Hochschuldidaktik (14)
- Ausbildung (13)
- machine learning (13)
- Cloud Computing (11)
- answer set programming (11)
- cloud computing (11)
- Hasso-Plattner-Institut (10)
- Hasso Plattner Institute (9)
Institute
- Institut für Informatik und Computational Science (207)
- Hasso-Plattner-Institut für Digital Engineering gGmbH (168)
- Hasso-Plattner-Institut für Digital Engineering GmbH (102)
- Extern (62)
- Mathematisch-Naturwissenschaftliche Fakultät (24)
- Institut für Mathematik (10)
- Digital Engineering Fakultät (7)
- Wirtschaftswissenschaften (7)
- Institut für Umweltwissenschaften und Geographie (5)
- Institut für Physik und Astronomie (4)
The constantly growing capacity of reconfigurable devices allows simultaneous execution of complex applications on those devices. The mere diversity of applications deems it impossible to design an interconnection network matching the requirements of every possible application perfectly, leading to suboptimal performance in many cases. However, the architecture of the interconnection network is not the only aspect affecting performance of communication. The resource manager places applications on the device and therefore influences latency between communicating partners and overall network load. Communication protocols affect performance by introducing data and processing overhead putting higher load on the network and increasing resource demand. Approaching communication holistically not only considers the architecture of the interconnect, but communication-aware resource management, communication protocols and resource usage just as well. Incorporation of different parts of a reconfigurable system during design- and runtime and optimizing them with respect to communication demand results in more resource efficient communication. Extensive evaluation shows enhanced performance and flexibility, if communication on reconfigurable devices is regarded in a holistic fashion.
1 Einleitung 1.1 Motivation 1.2 Aufgabenstellung 1.3 Aufbau der Arbeit 2 Fachliches Umfeld 2.1 Grid Computing 2.2 Idle Time Computing 3 Ressourcenpartitionierung 3.1 Ressourcenpartitionierung und Scheduling 3.2 Ressourcenpartitionierung in Idle Time Computing 3.2.1 Administrative Kontrolle der Ressourcen 3.2.2 Mindestgarantien zur Sicherstellung der Lauffähigkeit 3.3 Vorhandene Lösungen und verwandte Arbeiten 3.3.3 Ressourcenmanagement im Globus Toolkit 3.3.4 Ressourcenmanagement in Condor 3.3.5 Das GARA Framework 3.3.6 Distributed Resource Management Application API 3.3.7 Grid Resource Allocation Agreement Protocol 3.3.8 SNAP 3.3.9 OGSI-Agreement 3.3.10 PBS/Maui und andere Batch Systeme 3.3.11 Wide Area Distributed Computing 3.3.12 Weitere verwandte Arbeiten 3.3.13 Überlegungen zum Ressourcenbedarf 4 Ressourcenkontrolle in Desktopbetriebssystemen 4.1 Ressourcen 4.2 Ressourcenpartitionierung unter Linux 4.2.14 Festplattenkapazität 4.2.15 Arbeitsspeicher 4.2.16 Netzwerkbandbreite 4.2.17 CPU Kapazität 4.3 Ressourcenpartitionierung unter Microsoft Windows XP 4.3.18 Festplattenkapazität 4.3.19 Arbeitsspeicher 4.3.20 Netzwerkbandbreite 4.3.21 CPU Kapazität 4.4 Fazit 5 Entwurf und Design des Frameworks 5.1 Entwurfsgrundlage - Komponentenarchitektur 5.2 Architektur 5.2.22 Broker Server 5.2.23 Broker Software auf den Clients 5.2.24 Schnittstellen 5.3 Komponententypmodell 5.4 Ressourcenidentifikation und Ressourcenzuordnung 5.5 Anbindung ans Grid 5.6 Datenbankentwurf 5.7 XML RPC Schnittstelle 6 Implementierung 6.1 Broker Server 6.1.25 Datenbank 6.1.26 Komponenten 6.1.27 Webserverskripte 6.1.28 Database Crawler 6.2 Komponenten 6.2.29 Network 6.2.30 DSCP 6.2.31 Quota 6.2.32 FSF 6.3 Linux Client 6.3.33 Broker Client 6.3.34 Komponenten 6.4 Windows Client 6.5 Abhängigkeiten 7 Evaluierung 7.1 Durchgeführte Test- und Anwendungsfälle 7.1.35 Test der Clientsoftware 7.1.36 Test der Serversoftware 7.1.37 Durchführbare Anwendungsfälle 7.2 Evaluierung der Frameworkimplementierung 7.2.38 Performanz der Serverimplementierung 7.2.39 Zuverlässigkeit der Partitionierungen 7.3 Evaluierung von Traffic Shaping mit iproute2 7.3.40 Szenario 1 7.3.41 Szenario 2 7.3.42 Szenario 3 7.3.43 Fazit 8 Zusammenfassung und Ausblick 8.1 Fazit 8.2 Weiterentwicklung 8.2.44 Weiterentwicklungen auf Entwurfsebene 8.2.45 Weiterentwicklungen auf Implementierungsebene Anhang A: Details zum Datenbankentwurf Anhang B: Bildschirmfotos der Weboberfläche Anhang C: Quellcode Linux Broker Client Anhang D: Inhalt des beiliegenden Datenträgers
A common feature in Answer Set Programming is the use of a second negation, stronger than default negation and sometimes called explicit, strong or classical negation. This explicit negation is normally used in front of atoms, rather than allowing its use as a regular operator. In this paper we consider the arbitrary combination of explicit negation with nested expressions, as those defined by Lifschitz, Tang and Turner. We extend the concept of reduct for this new syntax and then prove that it can be captured by an extension of Equilibrium Logic with this second negation. We study some properties of this variant and compare to the already known combination of Equilibrium Logic with Nelson's strong negation.
The Runge-Kutta type regularization method was recently proposed as a potent tool for the iterative solution of nonlinear ill-posed problems. In this paper we analyze the applicability of this regularization method for solving inverse problems arising in atmospheric remote sensing, particularly for the retrieval of spheroidal particle distribution. Our numerical simulations reveal that the Runge-Kutta type regularization method is able to retrieve two-dimensional particle distributions using optical backscatter and extinction coefficient profiles, as well as depolarization information.
SandBlocks
(2020)
Visuelle Programmiersprachen werden heutzutage zugunsten textueller Programmiersprachen nahezu nicht verwendet, obwohl visuelle Programmiersprachen einige Vorteile bieten. Diese reichen von der Vermeidung von Syntaxfehlern, über die Nutzung konkreter domänenspezifischer Notation bis hin zu besserer Lesbarkeit und Wartbarkeit des Programms. Trotzdem greifen professionelle Softwareentwickler nahezu ausschließlich auf textuelle Programmiersprachen zurück.
Damit Entwickler diese Vorteile visueller Programmiersprachen nutzen können, aber trotzdem nicht auf die ihnen bekannten textuellen Programmiersprachen verzichten müssen, gibt es die Idee, textuelle und visuelle Programmelemente gemeinsam in einer Programmiersprache nutzbar zu machen. Damit ist dem Entwickler überlassen wann und wie er visuelle Elemente in seinem Programmcode verwendet.
Diese Arbeit stellt das SandBlocks-Framework vor, das diese gemeinsame Nutzung visueller und textueller Programmelemente ermöglicht. Neben einer Auswertung visueller Programmiersprachen, zeigt es die technische Integration visueller Programmelemente in das Squeak/Smalltalk-System auf, gibt Einblicke in die Umsetzung und Verwendung in Live-Programmiersystemen und diskutiert ihre Verwendung in unterschiedlichen Domänen.
Cloud computing is a model for enabling on-demand access to a shared pool of computing resources. With virtually limitless on-demand resources, a cloud environment enables the hosted Internet application to quickly cope when there is an increase in the workload. However, the overhead of provisioning resources exposes the Internet application to periods of under-provisioning and performance degradation. Moreover, the performance interference, due to the consolidation in the cloud environment, complicates the performance management of the Internet applications. In this dissertation, we propose two approaches to mitigate the impact of the resources provisioning overhead. The first approach employs control theory to scale resources vertically and cope fast with workload. This approach assumes that the provider has knowledge and control over the platform running in the virtual machines (VMs), which limits it to Platform as a Service (PaaS) and Software as a Service (SaaS) providers. The second approach is a customer-side one that deals with the horizontal scalability in an Infrastructure as a Service (IaaS) model. It addresses the trade-off problem between cost and performance with a multi-goal optimization solution. This approach finds the scale thresholds that achieve the highest performance with the lowest increase in the cost. Moreover, the second approach employs a proposed time series forecasting algorithm to scale the application proactively and avoid under-utilization periods. Furthermore, to mitigate the interference impact on the Internet application performance, we developed a system which finds and eliminates the VMs suffering from performance interference. The developed system is a light-weight solution which does not imply provider involvement. To evaluate our approaches and the designed algorithms at large-scale level, we developed a simulator called (ScaleSim). In the simulator, we implemented scalability components acting as the scalability components of Amazon EC2. The current scalability implementation in Amazon EC2 is used as a reference point for evaluating the improvement in the scalable application performance. ScaleSim is fed with realistic models of the RUBiS benchmark extracted from the real environment. The workload is generated from the access logs of the 1998 world cup website. The results show that optimizing the scalability thresholds and adopting proactive scalability can mitigate 88% of the resources provisioning overhead impact with only a 9% increase in the cost.
Scalable compatibility for embedded real-time components via language progressive timed automata
(2013)
Die korrekte Komposition individuell entwickelter Komponenten von eingebetteten Realzeitsystemen ist eine Herausforderung, da neben funktionalen Eigenschaften auch nicht funktionale Eigenschaften berücksichtigt werden müssen. Ein Beispiel hierfür ist die Kompatibilität von Realzeiteigenschaften, welche eine entscheidende Rolle in eingebetteten Systemen spielen. Heutzutage wird die Kompatibilität derartiger Eigenschaften in einer aufwändigen Integrations- und Konfigurationstests am Ende des Entwicklungsprozesses geprüft, wobei diese Tests im schlechtesten Fall fehlschlagen. Aus diesem Grund wurde eine Zahl an formalen Verfahren Entwickelt, welche eine frühzeitige Analyse von Realzeiteigenschaften von Komponenten erlauben, sodass Inkompatibilitäten von Realzeiteigenschaften in späteren Phasen ausgeschlossen werden können. Existierenden Verfahren verlangen jedoch, dass eine Reihe von Bedingungen erfüllt sein muss, welche von realen Systemen nur schwer zu erfüllen sind, oder aber, die verwendeten Analyseverfahren skalieren nicht für größere Systeme. In dieser Arbeit wird ein Ansatz vorgestellt, welcher auf dem formalen Modell des Timed Automaton basiert und der keine Bedingungen verlangt, die von einem realen System nur schwer erfüllt werden können. Der in dieser Arbeit vorgestellte Ansatz enthält ein Framework, welches eine modulare Analyse erlaubt, bei der ausschließlich miteinender kommunizierende Komponenten paarweise überprüft werden müssen. Somit wird eine skalierbare Analyse von Realzeiteigenschaften ermöglicht, die keine Bedingungen verlangt, welche nur bedingt von realen Systemen erfüllt werden können.
Scalable data profiling
(2018)
Data profiling is the act of extracting structural metadata from datasets. Structural metadata, such as data dependencies and statistics, can support data management operations, such as data integration and data cleaning. Data management often is the most time-consuming activity in any data-related project. Its support is extremely valuable in our data-driven world, so that more time can be spent on the actual utilization of the data, e. g., building analytical models. In most scenarios, however, structural metadata is not given and must be extracted first. Therefore, efficient data profiling methods are highly desirable.
Data profiling is a computationally expensive problem; in fact, most dependency discovery problems entail search spaces that grow exponentially in the number of attributes. To this end, this thesis introduces novel discovery algorithms for various types of data dependencies – namely inclusion dependencies, conditional inclusion dependencies, partial functional dependencies, and partial unique column combinations – that considerably improve over state-of-the-art algorithms in terms of efficiency and that scale to datasets that cannot be processed by existing algorithms. The key to those improvements are not only algorithmic innovations, such as novel pruning rules or traversal strategies, but also algorithm designs tailored for distributed execution. While distributed data profiling has been mostly neglected by previous works, it is a logical consequence on the face of recent hardware trends and the computational hardness of dependency discovery.
To demonstrate the utility of data profiling for data management, this thesis furthermore presents Metacrate, a database for structural metadata. Its salient features are its flexible data model, the capability to integrate various kinds of structural metadata, and its rich metadata analytics library. We show how to perform a data anamnesis of unknown, complex datasets based on this technology. In particular, we describe in detail how to reconstruct the schemata and assess their quality as part of the data anamnesis.
The data profiling algorithms and Metacrate have been carefully implemented, integrated with the Metanome data profiling tool, and are available as free software. In that way, we intend to allow for easy repeatability of our research results and also provide them for actual usage in real-world data-related projects.
Die vorliegende Arbeit erörtert die Frage, wie Nachwuchs für das Informatikstudium nachhaltig gesichert werden kann. Dazu werden Befragungen unter Schülerinnen und Schülern (13-16 Jahre), sowie aktuelle Informatik-Schnupperangebote für Schülerinnen und Schüler an deutschsprachigen Hochschulen vorgestellt und untersucht. Diese Gegenüberstellung zeigt deutlich, dass die Angebote nur bedingt eine breite Zielgruppe ansprechen und dass weitere Formate und Inhalte notwendig sind, um Schülerinnen und Schüler frühzeitig und in voller Breite zu erreichen und für das Informatikstudium zu begeistern. Daraus wird abgeleitet, dass Missverständnisse und Probleme mit der Informatik im Schulkontext aufgegriffen werden müssen. Das vorgestellte Programm Schulbotschafter Informatik stellt einen möglichen Weg dar, um dies zu erreichen und übliche Schnupperangebote zu ergänzen.
Business processes are fundamental to the operations of a company. Each product manufactured and every service provided is the result of a series of actions that constitute a business process. Business process management is an organizational principle that makes the processes of a company explicit and offers capabilities to implement procedures, control their execution, analyze their performance, and improve them. Therefore, business processes are documented as process models that capture these actions and their execution ordering, and make them accessible to stakeholders. As these models are an essential knowledge asset, they need to be managed effectively. In particular, the discovery and reuse of existing knowledge becomes challenging in the light of companies maintaining hundreds and thousands of process models. In practice, searching process models has been solved only superficially by means of free-text search of process names and their descriptions. Scientific contributions are limited in their scope, as they either present measures for process similarity or elaborate on query languages to search for particular aspects. However, they fall short in addressing efficient search, the presentation of search results, and the support to reuse discovered models. This thesis presents a novel search method, where a query is expressed by an exemplary business process model that describes the behavior of a possible answer. This method builds upon a formal framework that captures and compares the behavior of process models by the execution ordering of actions. The framework contributes a conceptual notion of behavioral distance that quantifies commonalities and differences of a pair of process models, and enables process model search. Based on behavioral distances, a set of measures is proposed that evaluate the quality of a particular search result to guide the user in assessing the returned matches. A projection of behavioral aspects to a process model enables highlighting relevant fragments that led to a match and facilitates its reuse. The thesis further elaborates on two search techniques that provide concrete behavioral distance functions as an instantiation of the formal framework. Querying enables search with a notion of behavioral inclusion with regard to the query. In contrast, similarity search obtains process models that are similar to a query, even if the query is not precisely matched. For both techniques, indexes are presented that enable efficient search. Methods to evaluate the quality and performance of process model search are introduced and applied to the techniques of this thesis. They show good results with regard to human assessment and scalability in a practical setting.