Theses

On this page you will find the current list of offered theses. The list of topics has to be understood as a collection of potential topics that could be adapted and typically this list is incomplete and it is worth to get in tough with us. Best is to address an Email to the secretary asking for an appointment and best you add already if you are looking for a bachelor or master thesis and if you study electric engineering, Informationssystemtechnik, the CT master programme or computer science

Open Thesis


12.
default
Hauser, Christopher
CPU Scheduling for Virtual Machines
Bachelor Thesis
Institut für Organisation und Management von Informationssystemen, Universität Ulm,
May 2015
in preparation

Abstract: follow the link to find out more

11.
default
Schubert, Lutz
Methods for realising virtually shared memory
Mastersthesis, Diplomarbeit
Institut für Organisation und Management von Informationssystemen, Universität Ulm,
2014
in preparation

Tags: MA, LutzSchubert, DA

Abstract: Uniformed memory access (UMA), respectively shared memory is still the easiest way to program large scale, parallel systems. Due to the scaling impact, modern systems, including multi-core processors, tend less and less to support UMA by hardware. This thesis aims at investigating different mechanisms to realise a middleware that exposes a shared memory to the developer, though the underlying memory architecture is distributed. The Partitioned Global Address Space (PGAS - http://www.pgas.org/) model is such an approach. The methods should be prototyped and evaluated on different systems. Some knowledge in parallel programming is recommended

10.
default
Schubert, Lutz
Measuring the OS performance impact factors
Mastersthesis, Diplomarbeit
Institut für Organisation und Management von Informationssystemen, Universität Ulm,
2014
in preparation

Tags: MA, LutzSchubert, DA

Abstract: Traditional (monolithic) operating systems are designed for x86 single-core processors, whereas modern processors incorporate more and more cores with a wide variety of co-processors. The goal of this thesis is to assess how well traditional operating systems interact with multi-core processors. The thesis should assess e.g. how cache-consistency protocols interact with data maintained by the operating system etc. Programming experience and some processor architecture knowledge are highly recommended.

9.
default
Schubert, Lutz
Controlling robots with mobile phones
Bachelorthesis, Projektarbeit
Institut für Organisation und Management von Informationssystemen, Universität Ulm,
2014
in preparation

Tags: BA, LutzSchubert, PA

Abstract: The goal of this thesis is to remote control simple robots, such as the Arduino Robot (http://arduino.cc/en/Main/Robot) using the sensors and communication of the mobile phone, e.g. tilting the smartphone to make the robot move into one direction. The program could be developed for either Android, WP and / or iOS. Programming knowledge is required.

8.
default
Schubert, Lutz
Distributed benchmarks for mobile phones
Bachelorthesis, Projektarbeit
Institut für Organisation und Management von Informationssystemen, Universität Ulm,
2014
in preparation

Tags: BA, LutzSchubert, PA

Abstract: Within this Bachelor thesis, typical benchmarks for distributed and in particular high performance computing systems (see e.g. http://icl.cs.utk.edu/hpcc/) shall be ported to smartphone(s). Basing on different communication modes (WLAN, Bluetooth etc.), the relative performance of the processors and network shall be evaluated. The benchmarks could be generated for either Android, WP and / or iOS. Programming knowledge is required.

7.
default
Wesner, Stefan
Integrating SDN Support in the IOMI OpenStack Testbed
Bachelorthesis, Mastersthesis,Projektarbeit
Institut für Organisation und Management von Informationssystemen, Universität Ulm,
2014
in preparation

Tags: MA, StefanWesner, JoergDomaschka, PA

Abstract: There is support from the OpenDaylight Project for OpenStack and a couple of “plugins” exist to integrate the cloud scheduler and open daylight ** Overview of available extensions and modules to add SDN support in OpenStack and analysis of their pros- and cons. ** Selection of the most appropriate technological options and integration of them within the IOMI OpenStack testbed ** Validation of the environment running a mix of scenarios and recording of the corresponding network setup such as - VM Movement and corresponding update of the Network Settings on the SDN Switches - VM Failure and Restart - Extension of an application by adding additional VMs - Shrinking of an application by stopping VMs ** Analyse and visualize the recorded network updates by extracting the necessary data from the Switch Rules and Table information in the SDN Switches before and after OpenStack triggered actions (der Teil übersteigt vmtl. den Umfang einer Bachelorarbeit je nachdem wie komplex diese Sache ist) * Vorbedingung: - Running Open Stack Testbed mit SDN fähigen Switchen

6.
default
Wesner, Stefan
Job Scheduling on Cluster Systems with SDN/OpenSM support
Bachelorthesis, Projektarbeit
Institut für Organisation und Management von Informationssystemen, Universität Ulm,
2014
in preparation

Tags: MA, StefanWesner, JoergDomaschka, PA

Abstract: Cluster systems are typically operated using queuing systems. User submit so called jobs typically consisting out of a shell script that collect input data from a shared storage device, launch the applications and move the result data from a temporary storage area to a permanent storage area. Cluster operators have the possibility to add on top of the user defined script so called “job prologues” allowing to execute operations before and just after the user job script. They are typically used to create local/job wide temporal storage areas and to perform clean-up operations. ** Overview of available queuing system solutions and their SDN support ** Selection of an Open Source queuing system such as OpenPBS as basis for integrating SDN/OpenSM support ** Implementation/Extension of job prologues setting up a “private” network environment on a per job basis and performing corresponding clean-up operations using SDN/OpenSM ** Validation of the environment running a mix of simulation jobs with the system ** Testing of the “sandbox” property by running jobs that aim to reach for neighbors nodes systematically as part of the submitted job mix

5.
default
Wesner, Stefan
Multi-Tenancy Network architecture for Cluster Systems
Masterthesis, Diplomarbeit
Institut für Organisation und Management von Informationssystemen, Universität Ulm,
2014
in preparation

Tags: MA, StefanWesner, JoergDomaschka

Abstract: Software Defined Networking has emerged as standardized mechanism to realize Multi-Tenancy in data centre architectures based on Ethernet connectivity. Due to its better suitability for scalable applications Infinband is still the dominating interconnect technology for cluster systems in the technical and high performance computing domain. ** Comparison on conceptual level of SDN/OpenFlow approaches and Infinband Subnet Manager ** Analysis of suitability for realizing a multi-tenancy technical computing/HPC System ** Overview of literature in this field and existing open source initiatives (e.g. OpenDaylight or OpenSM for Infiniband) ** Design and set-up of a test environment (simple beowulf style cluster with 8 nodes) and realizing a simple framework that allocate dynamically parts of the cluster to different customers - Option 1 is to use a very simple “job database” following a Producer/Consumer pattern that is just operated in FIFO manner - Option 2 is to use a queuing system such as OpenPBS and to make the network stuff in the prolog of the job script (maybe too complex?) - Option 3 is to use a Cloud Orchestrator to distribute the jobs (maybe too complex?) ** Some automated validation by realizing jobs running on the cluster that attempt to access nodes from other customers… Pre-Condition/Open Points: * SDN capable Switch with 1GE ~3000€+ * 8 “Nodes” * Infiniband components that support OpenSM (old bwGrid nodes?), costs for new IB Switch and IB cards with OpenSM support might be too pricy.

4.
default
Wesner, Stefan
Design and Realisation of an energy- and cost-efficient camera module supporting lecture recordings
Institut für Organisation und Management von Informationssystemen, Universität Ulm,
2014
in preparation

Abstract: The goal of this bachelor thesis is at least one reference design of a video capture platform that can be integrated with Opencast Matterhorn on basis of cheap micro controller or unix based computer platform. The platform must be able to record video in sufficient quality within different type of lecture halls. Part of the work is the prototypical realization and experimental validation of the design.

3.
default
Domaschka, Jörg
Comparing Distributed Event Processing Frameworks
Bachelorarbeit, Projektarbeit
Institut für Organisation und Management von Informationssystemen, Universität Ulm,
2014
in preparation

Tags: BA, PA, JoergDomaschka

Abstract: The need for processing huge numbers of events in a distributed system has led to the evolution of multiple distributed event processing engines such as S4 (Yahoo), Storm (Twitter), Riehmann and Vert.igo. The conceptual goals of this thesis are (a) to identify typical use cases for distributed event processing, (b) to classify the use cases based on common requirements and patterns, (c) analyse how existing distributed event processing engines support the identified categories. The practical part of this thesis demands a prototypical implementation of some use cases for various engines and a comparative evaluation.

Note: Die Notwendigkeit eine große Anzahl von Events in einem verteilten System verarbeiten zu müssen hat zu der Entstehung einiger Verteilter Event Prozessoren geführt, wie zum Beispiel S4 (Yahoo), Storm (Twitter), Riehmann oder Vert.igo. In dieser Arbeit sollen zunächst typische Anwendungsfälle für verteiltes Event Processing untersucht werden. Anschließend sollen Kategorien definiert werden, die die Unterstützung von Anwendungsfällen erfasst. Darauf aufbauend sollen verschiedene Frameworks untersucht werden und in diese Kategorien eingeordnet werden. Schließlich sollen einige typische Anwendungsfälle in mehreren Frameworks umgesetzt und schließlich vergleichend evaluiert werden.

2.
default
Domaschka, Jörg
Fehlertoleranz für die Esper CEP-Engine
Masterarbeit, Diplomarbeit
Institut für Organisation und Management von Informationssystemen, Universität Ulm,
2014
in preparation

Tags: MA, DA, JoergDomaschka

Abstract: Complex Event Processing (CEP) beschäftigt sich mit der Erkennung, Analyse, Gruppierung und Verarbeitung voneinander abhängiger Events. Werden CEP Engines in verteilten Systemen eingesetzt müssen sie mit der Fehleranfälligkeit verteilter Systeme zurechtkommen. In dieser Arbeit soll die Esper OpenSource CEP Engine auf ihre Einsatzfähigkeit in verteilten System hin untersucht werden. Daneben soll eine exemplarische Implementierung von Fehlertoleranzmechanismen stattfinden.

1.
default
Domaschka, Jörg
Integration des Virtual Nodes Frameworks in ein existierendes Middleware-System
Bachelorarbeit
Institut für Organisation und Management von Informationssystemen, Universität Ulm,
2014
in preparation

Tags: BA, PA, JoergDomaschka

Abstract: Das an der Universität Ulm entwickelte Virtual Nodes Framework vereinfacht die Implementierung fehlertoleranter Anwendungen erheblich. Es kapselt essentielle Funktionen einer fehlertoleranten Anwendung, wie zum Beispiel Gruppenmitgliedschaft und Ausfallerkennung in eigenen, austauschbaren Komponenten. Daneben ist es in der Lage Fehlertoleranzmechanismen so anzuwenden, dass die behandelte Anwendung nicht verändert werden muss. Darüberhinaus wurde das Framework so entworfen und implementiert, dass seine Logik unabhängig von einer eventuell eingesetzten Middleware ist. Zudem kann es relativ leicht in existierende Middleware Systeme integriert werden. Bisher existieren Virtual Nodes Anbindungen für CORBA, Java RMI sowie die Distributed XtreemOS Infrastructure. Um die Flexibilität des Virtual Nodes Frameworks zu demonstrieren, soll es in weitere Middleware Systeme und Infrastrukturen integriert werden. Jede dieser Integrationen kann als eigene Arbeit angegangen werden. Als Ziel-Systeme kommen zum Beispiel in Frage: Cloud Foundry, Tomcat Clustering, PaaSage Executionware, COSCA. Eine Integration in weitere Systeme ist auf Anfrage möglich.