thesis database system

Librarians/Admins

• EBSCOhost Collection Manager
• EBSCO Experience Manager
• EBSCO Connect
• Start your research
• EBSCO Mobile App

Clinical Decisions Users

• DynaMed Decisions
• Dynamic Health
• Waiting Rooms
• NoveList Blog

EBSCO Open Dissertations

EBSCO Open Dissertations makes electronic theses and dissertations (ETDs) more accessible to researchers worldwide. The free portal is designed to benefit universities and their students and make ETDs more discoverable. 

Increasing Discovery & Usage of ETD Research

EBSCO Open Dissertations is a collaboration between EBSCO and BiblioLabs to increase traffic and discoverability of ETD research. You can join the movement and add your theses and dissertations to the database, making them freely available to researchers everywhere while increasing traffic to your institutional repository. 

EBSCO Open Dissertations extends the work started in 2014, when EBSCO and the H.W. Wilson Foundation created American Doctoral Dissertations which contained indexing from the H.W. Wilson print publication, Doctoral Dissertations Accepted by American Universities, 1933-1955. In 2015, the H.W. Wilson Foundation agreed to support the expansion of the scope of the American Doctoral Dissertations database to include records for dissertations and theses from 1955 to the present.

How Does EBSCO Open Dissertations Work?

Your ETD metadata is harvested via OAI and integrated into EBSCO’s platform, where pointers send traffic to your IR.

EBSCO integrates this data into their current subscriber environments and makes the data available on the open web via opendissertations.org .

You might also be interested in:

Open Access Theses and Dissertations

Thursday, April 18, 8:20am (EDT): Searching is temporarily offline. We apologize for the inconvenience and are working to bring searching back up as quickly as possible.

Advanced research and scholarship. Theses and dissertations, free to find, free to use.

Advanced search options

Browse by author name (“Author name starts with…”).

Find ETDs with:

Written in any language English Portuguese French German Spanish Swedish Lithuanian Dutch Italian Chinese Finnish Greek Published in any country US or Canada Argentina Australia Austria Belgium Bolivia Brazil Canada Chile China Colombia Czech Republic Denmark Estonia Finland France Germany Greece Hong Kong Hungary Iceland India Indonesia Ireland Italy Japan Latvia Lithuania Malaysia Mexico Netherlands New Zealand Norway Peru Portugal Russia Singapore South Africa South Korea Spain Sweden Switzerland Taiwan Thailand UK US Earliest date Latest date

Sorted by Relevance Author University Date

Only ETDs with Creative Commons licenses

Results per page: 30 60 100

October 3, 2022. OATD is dealing with a number of misbehaved crawlers and robots, and is currently taking some steps to minimize their impact on the system. This may require you to click through some security screen. Our apologies for any inconvenience.

Recent Additions

See all of this week’s new additions.

About OATD.org

OATD.org aims to be the best possible resource for finding open access graduate theses and dissertations published around the world. Metadata (information about the theses) comes from over 1100 colleges, universities, and research institutions . OATD currently indexes 6,912,508 theses and dissertations.

About OATD (our FAQ) .

Visual OATD.org

We’re happy to present several data visualizations to give an overall sense of the OATD.org collection by county of publication, language, and field of study.

You may also want to consult these sites to search for other theses:

• Google Scholar
• NDLTD , the Networked Digital Library of Theses and Dissertations. NDLTD provides information and a search engine for electronic theses and dissertations (ETDs), whether they are open access or not.
• Proquest Theses and Dissertations (PQDT), a database of dissertations and theses, whether they were published electronically or in print, and mostly available for purchase. Access to PQDT may be limited; consult your local library for access information.

Global ETD Search

Search the 6,483,091 electronic theses and dissertations contained in the NDLTD archive:

The archive supports advanced filtering and boolean search.

EBSCO Open Dissertations

Search millions of electronic theses and dissertations (etds).

With EBSCO Open Dissertations, institutions and students are offered an innovative approach to driving additional traffic to ETDs in institutional repositories. Our goal is to help make their students’ theses and dissertations as widely visible and cited as possible.

This approach extends the work started in 2014, when EBSCO and the H.W. Wilson Foundation created American Doctoral Dissertations which contained indexing from the H.W. Wilson print publication, Doctoral Dissertations Accepted by American Universities, 1933-1955. In 2015, the H.W. Wilson Foundation agreed to support the expansion of the scope of the American Doctoral Dissertations database to include records for dissertations and theses from 1955 to the present.

Get involved in the EBSCO Open Dissertations project and make your electronic theses and dissertations freely available to researchers everywhere. Please contact Margaret Richter for more information.

ProQuest Dissertations & Theses Global

• What is ProQuest Dissertations & Theses Global
• ProQuest Dissertations eLearning Companions

E-Learning Modules

Webinar recordings, powerpoint presentations, additional resources, support center articles.

• Unique Features

Product Access & Information

• Additional Language Resources
• ProQuest Support Center
• Platform Status Page
• Upcoming Webinars

ProQuest Dissertations & Theses Global   is a wealth of unique global scholarship, which is a credible and quality source to Uncover the Undiscovered research insights and intelligence in easiest and most effective ways. The equitable discoverability of more than 5.8 million dissertations and theses with coverage from year 1637, allows researchers to amplify diverse voices and place their research in a global context. The database offers nearly 3.2 million full texts for most of the dissertations added since 1997.

By leveraging the rich citation data found in ProQuest Dissertations & Theses Global and with new citation insight tool, researchers can benefit from focused pathways of discovery to build foundational knowledge on various research topics. Over 200,000 new dissertations and theses are added to the database each year to enrich the citation data continuously.

For more information about the ProQuest Dissertations & Theses Global , navigate to the Content Page .

ProQuest Dissertations & Theses Global   Database  is also part of ProQuest One Academic .  ProQuest Dissertations & Theses Global resides on the ProQuest Platform. For more information about the ProQuest Platform search and display features, see the  ProQuest Platform LibGuide .

The Dissertations Bootcamp eLearning Modules are a free resource that help support graduate student planning, writing, and research.

ProQuest Dissertations and Theses for the Student, Citation Connections

Here you can have a preview of the new features just launched for the Cited Reference documents in ProQuest Dissertations & Theses Global.

ProQuest Dissertations and Theses for the Librarian

Intended for Librarians who want to learn how to use the database's advanced search to support subject area research at their institution. Duration: 2 minutes.

ProQuest Dissertations and Theses for the Student, Searching Titles and Languages

This session reviews how Students, both Masters or PhD, can use the database's advanced search to identify known dissertations by title and search/analyze by languages other than English. Duration: 3 minutes.

ProQuest Dissertations and Theses for the Student, Searching Names

This session reviews how Students, both Masters or Ph, can use the database's advanced search to identify dissertations of known Authors or Advisors and further refine/analyze them. Duration: 4 minutes.

ProQuest Dissertations and Theses for the Student, Cited References

This session reviews how Students, both Masters or PhD, can use the dissertations to retrieve and explore further the Cited References. Duration: 4 minutes.

ProQuest Dissertations and Theses for the Student, Supplemental Files

This session reviews how Students, both Masters or PhD, can identify dissertations with Supplemental files which may contain useful materials for their graduate work. Duration: 3.5 minutes.

ProQuest Dissertations and Theses for the Student, Subject Searching

This session will show Students, both Masters or PhD, some Search techniques both Basic and Advanced to locate dissertations on a certain topic. Duration: 5.5 minutes.

Webinar Title : Best Practices for Searching ProQuest Dissertations and Theses Global

This session demonstrates how users can utilize the best practices of searching the " ProQuest Dissertations and Theses Global  database" to connect with relevant information for their academic work. Duration:  52 minutes.

Materials in English - Figures (Database size) and Platform features images now updated as of March 2023

• PQDT Global Basic Version PPT
• PQDT Global Advanced Version PPT
• PQDT Global Citation Connection PPT
• ProQuest ETD Dissemination Program
• ProQuest Dissertations & Theses Citation Index Training Assest: ProQuest Dissertations & Theses Citation Index PPT- May 2024
• Marketing Toolkit
• New! Embedded eLearning for Undergraduates
• Request Training
• Top Dissertations
• ProQuest Dissertations & Theses Citation Index
• FAQ- What’s New for PQDT
• ProQuest Dissertations & Theses Global Data Used in Research Projects
• New Change in ProQuest Dissertations & Theses Global (PQDT) Structure and its Impact on Usage Reports in 2023
• New Change in ProQuest Dissertations & Theses Global (PQDT) Structure and its Impact on Saved Searches & Alerts
• Best Practices for Incorporating ProQuest Dissertations & Theses Global Product into EBSCO Discovery
• Next: Content >>
• Last Updated: May 4, 2024 10:06 AM
• URL: https://proquest.libguides.com/pqdt

• Harvard Library
• Research Guides
• Faculty of Arts & Sciences Libraries

Computer Science Library Research Guide

Find dissertations and theses.

• Get Started
• How to get the full-text
• What is Peer Review?
• Find Books in the SEC Library This link opens in a new window
• Find Conference Proceedings
• Find Patents This link opens in a new window
• Find Standards
• Find Technical Reports
• Find Videos
• Ask a Librarian This link opens in a new window

Engineering Librarian

How to search for Harvard dissertations

• DASH , Digital Access to Scholarship at Harvard, is the university's central, open-access repository for the scholarly output of faculty and the broader research community at Harvard.  Most Ph.D. dissertations submitted from  March 2012 forward  are available online in DASH.
• Check HOLLIS, the Library Catalog, and refine your results by using the   Advanced Search   and limiting Resource  Type   to Dissertations
• Search the database  ProQuest Dissertations & Theses Global Don't hesitate to  Ask a Librarian  for assistance.

How to search for Non-Harvard dissertations

Library Database:

• ProQuest Dissertations & Theses Global

Free Resources:

• Many  universities  provide full-text access to their dissertations via a digital repository.  If you know the title of a particular dissertation or thesis, try doing a Google search.  

Related Sites

• Formatting Your Dissertation - GSAS
• Ph.D. Dissertation Submission  - FAS
• Empowering Students Before you Sign that Contract!  - Copyright at Harvard Library

Select Library Titles

• << Previous: Find Conference Proceedings
• Next: Find Patents >>
• Last Updated: Feb 27, 2024 1:52 PM
• URL: https://guides.library.harvard.edu/cs

Harvard University Digital Accessibility Policy

A-Z Databases: EThOS: Welcome

Content, coverage & description.

EThOS is a UK wide repository of digitized doctoral theses. It provides full text access to 250,000 theses published in the UK as well as bibliographic records of non digitized theses. A quick and free  registration is required to access the full text content.  

• EThOS e-Theses Online Service This link opens in a new window The British Library digital repository for UK research theses offering a central access point to UK doctoral theses. The majority of universities in the UK are members. You can cross-search over 500,000 theses including those available for immediate download. more... less... Not Searchable via One Search. You must first register on an individual basis with Ethos in order to make requests through the Ethos website. The Ethos site provides further information.

When to use

EThOS is an excellent resource for finding highly specialised and original research. S tudents embarking on a doctoral programme may wish to see examples of successful theses in their own discipline and identify areas of research that have not already been covered.  

Video guide: Using Ethos to find UK theses

University of Kent (2019) Using Ethos to find UK theses.  28 August. Available at: https://www.youtube.com/watch?v=n_c3ZRC5tG0 (Accessed: 13 July 2021).

Basic search

Keep search terms simple, e.g. 'organisational culture', 'business ethics' , ' eosinophilic inflammation', ' diabetes management'. 

Full text availability is indicated by a green open padlock symbol next to the search results.  You can restrict the search results to full text only by checking the box  ' limit search to items available for immediate download' below the search bar.

Search results are listed in order of relevance. You can change the order of results to A to Z by author or year (most recent to oldest and vice versa) by using the 'sort by" option above the search results. 

Click the relevant search result to view the abstract and download the thesis (if applicable). The full text can be accessed via the 'Immediate download'   link when this is available. You will be prompted for the username and password you chose when you registered with EThOS. 

Advanced Search

The advanced search tool enables you to search for a specific thesis by author and title. Use  the drop-down menus next to the search boxes to select your desired search criteria. 

You can also combine search terms using the Boolean operators 'AND', 'OR', 'NOT' (drop-down menu to the left of the search boxes). Example:

'higher education' (box one) AND  'social inequality' (box two)  will instruct  the database to look for theses which contain both search terms in the thesis abstracts or titles. 

'social inequality' (box one) OR  'poverty' (box two) will instruct the database to search for each search term separately. 

'higher education' (box one) AND  'social inequality' (box two) OR  'poverty' (box three) will instruct the database to search for articles which contain the first search term 'higher education' and either the second 'social inequality' or the third 'poverty'. 

Try to keep your search terms simple especially when combining search terms.  EThOS is a highly specialised database and over complicating searches may lead to a zero result outcome. 

Help and Support

Library staff are available to help you to use all of our online databases and electronic journal services. 

Contact us at  [email protected]  or via the  Self Service Port al . 

Copyright Statement

• Last Updated: Mar 24, 2022 1:40 PM
• URL: https://uws-uk.libguides.com/EThOS

• DSpace@MIT Home
• MIT Libraries

This collection of MIT Theses in DSpace contains selected theses and dissertations from all MIT departments. Please note that this is NOT a complete collection of MIT theses. To search all MIT theses, use MIT Libraries' catalog .

MIT's DSpace contains more than 58,000 theses completed at MIT dating as far back as the mid 1800's. Theses in this collection have been scanned by the MIT Libraries or submitted in electronic format by thesis authors. Since 2004 all new Masters and Ph.D. theses are scanned and added to this collection after degrees are awarded.

MIT Theses are openly available to all readers. Please share how this access affects or benefits you. Your story matters.

If you have questions about MIT theses in DSpace, [email protected] . See also Access & Availability Questions or About MIT Theses in DSpace .

If you are a recent MIT graduate, your thesis will be added to DSpace within 3-6 months after your graduation date. Please email [email protected] with any questions.

Permissions

MIT Theses may be protected by copyright. Please refer to the MIT Libraries Permissions Policy for permission information. Note that the copyright holder for most MIT theses is identified on the title page of the thesis.

Theses by Department

• Comparative Media Studies
• Computation for Design and Optimization
• Computational and Systems Biology
• Department of Aeronautics and Astronautics
• Department of Architecture
• Department of Biological Engineering
• Department of Biology
• Department of Brain and Cognitive Sciences
• Department of Chemical Engineering
• Department of Chemistry
• Department of Civil and Environmental Engineering
• Department of Earth, Atmospheric, and Planetary Sciences
• Department of Economics
• Department of Electrical Engineering and Computer Sciences
• Department of Humanities
• Department of Linguistics and Philosophy
• Department of Materials Science and Engineering
• Department of Mathematics
• Department of Mechanical Engineering
• Department of Nuclear Science and Engineering
• Department of Ocean Engineering
• Department of Physics
• Department of Political Science
• Department of Urban Studies and Planning
• Engineering Systems Division
• Harvard-MIT Program of Health Sciences and Technology
• Institute for Data, Systems, and Society
• Media Arts & Sciences
• Operations Research Center
• Program in Real Estate Development
• Program in Writing and Humanistic Studies
• Science, Technology & Society
• Science Writing
• Sloan School of Management
• Supply Chain Management
• System Design & Management
• Technology and Policy Program

Collections in this community

Doctoral theses, graduate theses, undergraduate theses, recent submissions.

The properties of amorphous and microcrystalline Ni - Nb alloys. 

Towards Biologically Plausible Deep Neural Networks 

Randomized Data Structures: New Perspectives and Hidden Surprises 

EThOS: e-theses online service Open access EThOS: e-theses online service Open access

EThOS, provided by the British Library, is a free online service providing access to UK doctoral theses.

Access EThOS is an open access resource.

Content  EThOS, provided by the British Library, is a free online service providing access to UK doctoral theses. It does not cover MPhils or master's dissertations.

EThOS aims to provide a central listing of all doctoral theses awarded by UK higher education institutions, with the full text of as many theses as possible. 

The database includes more than 600,000 records. Around 4,000 law theses are covered, dating from the 1920s to the present day.

Searching  EThOS has basic and advanced search facilities. Searches can be limited to theses available for immediate download.

Advanced search allows users to search by author, title, awarding body, year of award and other criteria. Boolean connectors (AND, OR, AND NOT) can be selected from a drop-down menu

Downloading  Many theses are available for download; it is necessary to create a free account to do this. 

There is an option to request digitisation of a thesis if it is not yet available for download. Sometimes this is free, but sometimes there is a charge (see FAQs). 

Help A Help menu and an FAQ page are available.

HKUST Electronic Theses Database: Submit E-thesis

• Submit E-thesis
• Find Theses

HKUST Theses Submission

The Library collects all HKUST theses and dissertations in  electronic format.

The PDF copy will be loaded into the

• HKUST Electronic Theses Database
• HKUST SPD | The Institutional Repository  
• sciences & engineering
• humanities & social sciences

Thesis Submission Requirement & Procedure

Requirement:

On successful completion of the thesis examination, students must submit:

• 1  soft copy  of your approved  thesis (PDF) to Library. Be sure to follow the HKUST Guidelines on Thesis Preparation . 
• 1 soft copy of the signed Signature Page and the Authorization Page (PDF) from your department

Submission Procedures:

• your approved thesis (PDF)
• signed signature page (authentic signature, digital signature, and signature chop are accepted) & signed authorization page (PDF)
• PhD students: Check the agreement to deposit the thesis to ProQuest. You will agree to the Library, on your behalf, to submit the thesis and abstract to the world-recognized ProQuest Dissertation database for publication.

E-thesis PDF File Checking and Printing Service

This simple application will double check your thesis for simple errors that require your attention before submission. It can catch even the smallest error that will prevent full submission (e.g., slight margin errors or excessive blank spaces)

Download the application from MTPC . ( Video Guide )

• Username: e_thesis_app
• Password: pgso2015

For additional assistance and troubleshooting, refer to  Creating PDF files for E-thesis Submission Guide   Section F .

Direct inquiries to Mr Joe Wong ( ptjoe @ust.hk)

Q: What are the requirements for thesis submission?

A: For thesis preparation and formatting, please visit Guidelines on Thesis Preparation for more details.

Q: Is there any format restriction for images or videos in HKUST Electronic Thesis Submission?

A: The HKUST Electronic Thesis Submission Form accepts all kinds of image or video format you provide with unlimited file size. However, please note that you may encounter problems uploading extremely large files due to network problems or other issues.

Q: Who holds the copyright of theses after the submission?

A: According to the University’s Intellectual Property Policy , students hold sole copyright ownership of their thesis. More details about copyright are available at the Copyright Libguide .

Q: When will my thesis be released to the public?

A: If you did not send a request to keep the thesis confidential, your thesis will be released once all the library processes are completed. Otherwise, the thesis will be released when the confidential period ends.

Get Assistance

About Thesis Submission

[email protected]

Additional Contact Details

• Next: Find Theses >>
• Last Updated: May 10, 2024 7:34 PM
• URL: https://libguides.hkust.edu.hk/e-thesis

• Renew Books
• Hours and Locations

You are here

Penn state electronic theses and dissertations.

Provides access to Penn State electronic theses and dissertations.

Students! We are looking forward to work with you on the open topics listed below. You may also contact us if you think that you have identified an interesting challenge that may fit the research profile of our chair.

In either case, please send e-mail to [email protected] (German is fine). Make sure to include the following information:

• Which of our chair’s courses did you attend?
• Beyond database systems, which courses/seminars have particularly raised your interest?
• Are you a hands-on hacker type or do you prefer working with pencil and paper ?
• Which programming languages are part of your toolbox? Which of these do you master ?
• Did you work with functional programming languages (beyond the bits of Scheme taught in the introductory CS courses)?
• Any projects you have worked on in the recent past?
• What would be a suitable date to start the work on your thesis? What are your other commitments (e.g., courses, exams, day job) in the thesis time frame?
• Please also enclose your transcript of records.

In Progress

Ludwig Kolesch (B.Sc.)

Bringing Flummi to WITH MUTUALLY RECURSIVE on Materialize DB

Our Flummi compiler currently emits SQL code based on WITH RECURSIVE. Aim of this bachelor thesis is to bring this code to Materialize DB using its specialized variant of recursive CTEs: WITH MUTUALLY RECURSIVE.

Contact: Tim Fischer

Markus Holder (B.Sc.)

Bringing Flummi to umbra.trampoline

Our Flummi compiler currently emits SQL code based on WITH RECURSIVE. Aim of this bachelor thesis is to bring this code to Umbra using its specific tool for iteration: umbra.trampoline().

Contact: Louisa Lambrecht • Tim Fischer

Fredo Hogen (B.Sc.)

How expressive is Flummi really? Can it run Advent of Code?

Nico Faden (B.Sc.)

Bringing ORDINALITY to DuckDB

Contact: Tim Fischer • Denis Hirn

Felix Kofink (M.Sc.)

Database-Backed Computation Over Weather Data

Contact: Tim Fischer • Torsten Grust

Björn Bamberg (M.Sc.)

DuckDB: Improved Support for List Functions

DuckDB is an open source in-process SQL OLAP database management system and comes with support for nested values such as arrays. However, there are certain features that are not implemented at this time. This thesis aims to identify these missing features, implement and test them, and ideally have them merged. This project requires knowledge of C++ development as well as knowledge of database systems internals (i.e., as discussed in DB2).

Contact: Denis Hirn • Torsten Grust

Xenia Wetzel (B.Sc.)

Writing an Interpreter for a Database-Coupled Language

Zora Pidde (B.Sc.)

Extending ByePy with geometric types and operators

In an effort to demonstrate the broader applicability of SQL as a compilation target our group, over the past year, has extended our original PL\SQL to SQL compiler with Byepy — a Python frontend. The currently support data types only include the very basics — integers, float, strings, booleans, composite types, and arrays. This selection already covers quite a few usecases, but many database system support fancier data types like dates, 2D shapes, JSON, etc.

Adrian Müller (M.Sc.)

Extending WITH RECURSIVE With Multiple Working Tables

The WITH RECURSIVE clause of SQL typically allows only a single working table. In this work, we will investigate what happens if we remove this restriction and allow any number of working tables to read from and write to. This project requires knowledge of C development as well as knowledge of database systems internals (i.e., as discussed in DB2).

Contact: Denis Hirn

Alexander Götz (B.Sc.)

From PL/pgSQL to C : What Are the Optimizations?

PL/pgSQL is an imperative language extension that is implemented as an interpreter layered on-top of PostgreSQL. This introduces friction during execution because for each embedded statement, an executor must be instantiated, run, and finally purged. In our research, we have developed methods to eliminate this problem by Compiling PL/SQL Away, which generates a single SQL query. In this thesis, we investigate what happens when we compile PL/pgSQL functions into equivalent C functions, thus eliminating the interpretation overhead of the PL/pgSQL interpreter.

Romain Carl (B.Sc.)

Implementing Abstract Machines in SQL

Abstract machines allow for detailed and precise analysis of how computer systems work. Abstract machines for functional programming languages include the SECD machine (1964) and the Krivine machine (1985). This thesis explores the implementation of these abstract machines in SQL using PostgreSQL and DuckDB.

Contact: Louisa Lambrecht

Lukas Sailer (M.Sc.)

Mapping GraphQL Queries to SQL

GraphQL (or: GQL) is a query language for APIs, offering the ability to replace inefficient request-by-request API calls by bulk-oriented request processing. GraphQL understands links between API resources and can follow such references. When these resources are held in a relational database, can a given GQL query be systematically translated into a single SQL query that traverses relationships and collects the requested resources? Can we find a translation from GQL to SQL that avoids the infamous n+1 problem when multi-valued references are traversed?

Contact: Torsten Grust

Tim Fischer (M.Sc.)

ByePy : Compilation of Python to SQL

Over the the last few years, our group has developed a compilation approach to compile PL/SQL UDFs to plain SQL queries. Post compilation, evaluation entirely happens on the SQL side of the fence. In particular, we use trampolined style to compile arbitrarily nested iterative control flow in PL/SQL into SQL’s recursive common table expressions. By applying the exact same compilation techniques to database-driven Python code, computations can be moved directly into the DBMS and thus close to the data they are working with.

Phil Elgert (M.Sc.)

Creating a Website with a Static Site Generator

Contact: Denis Hirn • Tim Fischer

Daniel Täsch (M.Sc.)

How to Optimize What Is Slow in Data Provenance and Why You Should Do It

In recent years, our group has developed a novel approach to provenance analysis for SQL. This approach is based on query rewriting. Given a query Q to be analyzed for its data provenance, the rewritten queries Q1 and Q2 are produced. Through evaluation of these two queries, data provenance is derived. The task of this Master thesis is to integrate one (or multiple) optimization steps in the query rewriting. Especially Q2 could benefit from static optimization of set expressions.

Contact: Tobias Müller

Marcus Huber (M.Sc.)

Optimization of PL/pgSQL Translations Using Batching

Our compilation approach compiles PL/SQL UDFs to plain SQL queries. We use trampolined style to compile arbitrarily nested iterative control flow in PL/SQL into SQL’s recursive common table expressions. Batched execution of function invocations as well as using multiple recursive anchors potentially improves performance of these translations. After manually compiling several examples, an experimental run-time evaluation should clarify in which cases batching improves performance.

Madeleine Mauz (M.Sc.)

Using Hashtables in Functional-Style UDF Compiled CTEs

Contact: Christian Duta

Felix Kofink (B.Sc.)

Constructing static websites based on AirTable databases

Thora Daneyko (B.Sc.)

Data Provenance for PL/pgSQL

Provenance in SQL describes the relationship between the output of a query and its input data (tables). In our recent research we employed an approach of language-level SQL rewriting to derive fine-grained where- and why- provenance for the output data of SQL SELECT-queries. [1] PL/pgSQL is a procedural language for PostgreSQL wich allows us to write functions using procedural control structures (LOOP, IF, EXCEPTION) on top of embedded SQL statements (SELECT-queries, but also INSERT, UPDATE and DELETE statements).

Contact: Benjamin Dietrich

Jonatan Braun (M.Sc.)

Extend SQLite3 with Support for LATERAL Join

SQLite3 is the most widely used server-less Database Engine today. SQLite3 supports a wide range of SQL features, but it is lacking LATERAL-Joins. The LATERAL-keyword allows to reference columns provided by preceding FROM items. The goal of this thesis is to extend SQLite3 with support for LATERAL joins. This project requires skills in C development and knowledge about database internals (i.e., as discussed in Datenbanksysteme 2).

Tobias Burghardt (M.Sc.)

From Recursion To Iteration: Compiling SQL UDFs with Continuations

The goal of this thesis is to investigate and evaluate recursion removal techniques for (PL/pg)SQL UDFs. Continuation-Passing-Style, Defunctionalization, and Trampolining-Style can be used to transform recursive programs into equivalent tail-recursive programs, and finally into iterative programs. After manually compiling several examples, an experimental run-time evaluation should clarify whether this strategy is advantageous over naïve evaluation.

Pascal Engel (B.Sc.)

How-Provenance Through Query Rewriting

In this BSc thesis, a SQL-to-SQL compiler is to be implemented. In recent research, we employed a two-step query rewrite strategy in order to compute Where- and Why-provenance for SQL. Using the approach from above, we want to implement a different provenance semantics, i.e. How-provenance. Related work by O’Grady et al. employed a query compilation approach to How-provenance. Originally, Green et al. introduced the notion of How-provenance in context of relational algebra and semirings.

Contact: Benjamin Dietrich • Tobias Müller • Torsten Grust

Fabian Bauer (M.Sc.)

Projektorientierter Informatikunterricht zum Thema “Relationale Datenbanksysteme”

Tim Fischer (B.Sc.)

Utilizing parallelism in the two-phase approach of translated fsUDFs

We translate fsUDFs, UDF with recursive calls, into SQL. The way the UDF then evaluates does not rely on recursive calls anymore. Instead, we first determine the call graph of the fsUDF top-down and then evaluate it bottom-up with the help of (standard SQL) recursive queries. This two-phased approach hides potential for parallelism which we want to explore in this work.

Martin Gabrich (M.Sc.)

Building a Turn Based Strategy Game with Database Backend

Jonas Wolff (B.Sc.)

Frontend for Creating Input for a Database-Supported Map Generator

To support our approach of building maps for video games completely within a database, the goal of this theses is a visual tool for creating building blocks for a map.

Peter Richter (M.Sc.)

How Does ToWithRecursive Hold Up?

Compare the ease of implementation, runtime and other features of ToWithRecursive in PostgreSQL with other analytic and framework environments.

Cedric Breuning (B.Sc.)

Implementierung von Program Slicing auf fsUDFs

Wir suchen nach einer geschickten Implementation von Program Slicing auf fsUDFs.

Stephan Biastoch (M.Sc.)

Intuitive Recursion Preprocessor for PostgreSQL

We aim to be able to write straight forward recursive user-defined functions functions (UDF) in PostgreSQL using LANGUAGE SQL. To achieve this, we need a preprocessor, translating a proper recursive UDF into something PostgreSQL can process utilising WITH RECURSIVE. This preprocessor and its rules have yet to be written and such encompasses the main work in this thesis.

Andreas Herzog (B.Sc.)

Normalization of SQL Queries

The analysis, transformation, and optimization of SQL queries can be streamlined and simplified when the incoming SQL queries show little (or: less) syntactic variety. This thesis will design a normal form for SQL queries that reduces the number of special cases, syntactic oddities, and countless abbreviations in SQL. The resulting SQL queries will be somewhat more verbose and explicit but exhibit syntactic regularity. Normalization is equivalence-preserving (but headache-reducing ;-). Based on our group’s Haskell-based parser for (Postgre)SQL queries, the student will implement the normalization transformation in Haskell.

Noah Doersing (M.Sc.)

Translating UnicodeMath into MathML

Allow the use of UnicodeMath — a Unicode-based intuitive rendering of complex math formulae — in Markdown documents. Translate UnicodeMath input into MathML for rendering (e.g., in Web browsers). Integrate this translator into Morgan McGuire’s Markdeep to make UnicodeMath immediately usable in day-to-day document authoring.

Jonas König (B.Sc.)

Evaluating Game Engines To Infuse Them With DB Technology

While games are at the surface mostly a playful distraction for relaxed afternoons, they deal with considerable amounts of data under the hood. This thesis evaluates the aptness of several open game frameworks for their use with database technology.

Contact: Daniel O'Grady

Louisa Lambrecht (B.Sc.)

Functional Universe

Games are an opportunity to introduce students to programming gently. This thesis provides an abstraction for a game so that students can easily implement their own game clients.

Alexander Mühlbauer (B.Sc.)

Iteration to Tail Recursion in Python

Certain iterative python functions can be transformed into tail recursive functions using static program analysis. In this thesis, we strive to formulate and implement an automatic transformation method. Our goal: Iterative function goes in, tail recursive function comes out.

Gabriel Paradzik (B.Sc.)

Language-Level Provenance Analysis of SQL

We want you to embed a variant of Data Provenance Analysis directly into SQL queries. The two-step analysis approach to be used has been described in recent publications. In previous research, we first compiled SQL into imperative code and afterwards ran the provenance analysis. Novel to this topic is to omit the compilation step. This means that your implementation will transform an input SQL query into two new SQL queries. While the first query produces execution logs, the second derives the actual data provenance (see Figure 4 in the paper mentioned above).

Contact: Benjamin Dietrich • Tobias Müller

Denis Hirn (M.Sc.)

PGcuckoo — Inject physical plans into PostgreSQL

Assemble physical execution plans for PostgreSQL and have them executed in the context of regular SQL queries. Creates the ability to design and compile such physical plans directly — no SQL or SQL compilation involved. Enables the design of PostgreSQL code generators that do not rely on SQL serialization and re-parsing. Building blocks: Haskell library and PostgreSQL extension.

Martin Fuß (B.Sc.)

Provenance of SQL Transactions

In our previous work we investigated and implemented Language-Level Provenance Analysis of SQL SELECT-queries. With this work we want to expand our two-phase approach to derive provenance of arbitrary SQL transactions (meaning sequences of SQL DML statements, including SELECT, INSERT, UPDATE and DELETE). The thesis is about the implementation of a SQL transformation tool. It returns two versions of a given SQL program: One equivalent, but with some additional logging. A second one, deriving the program’s provenance.

Zhili Zhang (M.Sc.)

Visualizing Query Plans

This thesis implements a visualiser for query plans in vanilla Javascript to be used in teaching.

Denis Hirn (B.Sc.)

Compilation of SQL into KL

This thesis is about implementing a compiler in Haskell. The source dialect of SQL consists of reading queries only. The compiler has to cover aggregations, correlated subqueries and window functions. The target language will be KL (short for: Kernel Language), an imperative language designed for provenance analysis. All input queries have type annotations. These are to be translated as well.

Contact: Daniel O'Grady • Tobias Müller • Torsten Grust

Moritz Bruder (M.Sc.)

Database-Supported Haskell: Efficient Code Generation for MonetDB

MonetDB is a highly-efficient column-store database system that is a promising execution platform for Database-Supported Haskell (DSH). The goal of this student project is to improve an existing code generator to unlock the full potential of MonetDB. Students will Analyze and profile the code that is currently produced Implement improvements to the current code generator Work with the MonetDB optimizer framework Get to know the internals of an innovative database system This project requires solid skills in Haskell and knowledge about database system internals.

Contact: Alexander Ulrich

Lena Weinmann (B.Sc.)

Design and Implementation of a Browser-Based User Interface for Habitat

Habitat is an observational debugger for SQL developed in our research group. Users can highlight parts of a buggy SQL query in order to observe intermediate results of the evaluation of particular subexpressions and thus learn about the query’s actual runtime behaviour. Debugging with Habitat is an interactive process where the user places new markings based on the debugger output of previous markings, to narrow down the precise location of an error cause, step by step; playfully hunting the bug.

Noah Doersing (B.Sc.)

Kernel Language to LLVM Compiler

Compiler für die Minimalsprache “Kernel Language” in die hardwarenahe LLVM-Umgebung.

Simon Poschenrieder (M.Sc.)

Reduce query delay in IBM IDAA (Neteeza)

Optimize transaction placement in the IDAA transaction queue to minimize query delay due to update latencies between DB2 and IDAA.

Jonas Lingg (B.Sc.)

Simple Haskell Graphics Library (based on JavaScript Canvas)

Build a simple Haskell-based graphics library, designed to be used in the classroom (target courses: Functional Programming or a future iteration of Informatik 1). Create a shallow wrapper around the Haskell package blank-canvas based on the JavaScipt Canvas API. Hide gory package details and create additional graphics primitives: Add turtle graphics (definitely needed) Add interactivity, e.g. mouse/keyboard event handling? (nice to have)

Martin Lutz (B.Sc.)

Visualization of How-Provenance

This is a topic for a Bachelor or Master thesis in context of our Data Provenance research project. In short, the research field of Data Provenance is about the computation of the exact origin of certain data pieces. In our case, we aim to analyse arbitrary SQL queries and return the exact tuple|cell locations of everything that went into the computation of the querie’s result. Our research approach does not utilize relational algebra which makes it different from most approaches found in the scientific literature.

Contact: Daniel O'Grady • Tobias Müller

Vanessa Rath (M.Sc.)

Analyse von Textkorpora im Bereich Recruiting

Marco Schneider (M.Sc.)

Clojure Library for Embedded Queries

Based on a Scheme prototype hacked by Mike Sperber. Maps PL objects to database tables in a flexible fashion. Maps query operations on the PL objects to database operations. Maps algebraic data types to relational tables. Attempts to target different DB backends (JSONiq?). Start work on actual MSc thesis in Mar 2016.

Jan Burchard (M.Sc.)

IBM Netezza Zone Maps and Geographic Data

Simulate and evaluate variants of IBM Netezza zone maps (index-like data structure that holds min/max values of designated columns) for multi-dimensional geographic data. In cooperation with IBM Research & Development, Germany (Böblingen)

Marco Häberle (B.Sc.)

Implementation of a Web-Based Frontend for JIT-Compiled PostgreSQL functions

Most of today’s database systems follow an interpreted query execution approach. Instead of generating a complete query tree, Just-in-time compilers empower us to compile parts of the tree to a intermediate language and finally to native machine code avoiding any interpretation overhead. These functions live in memory only making debugging as well as presenting the code difficult. Therefore a web-based frontend for the integrated display of query tree, LLVM intermediate language and x86 machine code has to be developed.

Contact: Dennis Butterstein

Daniel O'Grady (M.Sc.)

Optimized Supply Chain Management

Optimierung des Lagerbestandes auf der Basis von Bestell-/Verkaufsdaten der Vergangenheit. Kontext: SAP.

Hien Hung Nguyen (B.Sc.)

Peformance Optimizations for the PostgreSQL JIT Compiler

Jonas Weissensel (M.Sc.)

Ship route simulation based on location and weather data

Michael Zabka (M.Sc.)

Using Spark as a Backend for R

Explore the integration of database management systems and parallel data processing frameworks (here: Spark) into software environments for machine learning and statistical computing (here: R).

Stefan Burnicki (M.Sc.)

Design of a Portable SQL Wire Protocol

Develop a proxy that serializes SQL query result data into a common wire format (JSON?) and converts between data type representations—a single client library suffices to communicate with multiple RDBMS back-ends (Transbase, PostgreSQL). In cooperation with Grau Data (Stuttgart) and Transaction Software GmbH (Munich).

Nadejda Ismailova (B.Sc.)

Development of a Data Provenance Analysis Tool for Python Bytecode

Data Provenance describes the origins of data. In this thesis we propose an approach of computing data provenance when data of interest is created by a Python program and present a tool implementing this approach. In imple- menting the tool, we adopt the following idea. In the first step, the program is instrumented to log data. In the second step, the code is analyzed with respect to the logged data and provenance information is computed.

Peter Richter (B.Sc.)

Extract and Sanitize PostgreSQL Query Parse Trees

The majority of the research projects in our group analyze, transform, and compile SQL queries. These projects vary widely but all of them rely on an internal representation of the incoming SQL query text. PostgreSQL comes with a sophisticated SQL parser — that also incorporates a large number of semantic checks, type inference, and query simplifications/normalizations — and the output of this parser would be the ideal input for the research projects just mentioned.

Philipp Moers (B.Sc.)

Binoculars for Habitat: Implementation of a Web-Based Frontend for an Observational SQL-Debugger

Contact: Benjamin Dietrich • Torsten Grust

Tobias Fabritz (B.Sc.)

Evaluation of decimal arithmetic for predicate evaluation in LLVM

Janek Bettinger (B.Sc.)

Implementation of a Browser-based Interface for Provenance Analysis

Data provenance is a research topic dealing with the origins of generated data. An already existing tool by the Database Research Group of the University of Tübingen computes provenance relations of Python programs and outputs its results in the relational data format. The primary purpose of this thesis is the development of an interactive browser-based interface that makes the analysis results human-accessible.

Steffen Brennscheidt (B.Sc.)

Implementation of a DSH code generator for MonetDB5 MAL

Alexander Schiller (B.Sc.)

Implementation of an XML to Relational Data Format Conversion Tool

Viele Nicht-Informatiker möchten mit Daten arbeiten, die in XML Dumps enthalten sind. Um diese Daten in eine Form zu bringen, in der diese einfacher zu verwerten sind, haben wir ein Konvertierungstool geschrieben, das XML Dokumente in CSV Dateien umwandelt. Dabei ist das Programm weitgehend unabhängig vom verfügbaren Arbeitsspeicher, was es möglich macht auch sehr große XML Dumps zu konvertieren, und von den Programmierkenntnisse des Anwenders.

Simon Kalt (B.Sc.)

Instrumentation of Python Bytecode and Symbolic Program Analysis

As part of a project to help non-programmers understand computer programs, the goal of this work was to generate information on the behaviour of Python programs. The implementation described in this work uses an approach based on instrumenting Python bytecode to trace back variable assignments in order to build a graph depicting how the output of the program depends on its intermediate and input values. The presented version of the implementation enables the user to analyze simple programs by visualizing the data flow occurring during to the computation of a program’s output.

Development of a webservice frontend to support Performance-Simulation

Achim Kruse

GHC extensions for overloading list notation and rebinding of SQL-like monad comprehension notation

Tobias Müller

In Memory Implementation of Vector Primitives

DSH is a Haskell library for database supported program execution. It turns the database into a coprocessor for the Haskell runtime. Compilation of Haskell code for the database is a multi-step process. One of the intermediate languages is the vector algebra which is of interest to get a better understanding of the performance behaviour of DSH. We have implemented an in memory execution environment which executes vector algebra programs natively. The execution environment allows us to investigate the performance behaviour of each vector operation individually and without database context.

Alexander Zietlow

Performance analysis on database queries on IBM System Z, P & X

Worked on by Jessica Abele (BSc Bioinformatik) @ GE Healthcare, Munich

Representing Hierarchichal Structures in NoSQL Databases

Star wars database.

Sebastian Brandt

Use variant of XPath Accelerator to index nested PDF data

Implement data wrangler for dsh.

Database Supported Haskell (DSH) ist eine spezialisierte Abfragesprache für verschachtelte und sortierte Datensammlungen. Diese Ausarbeitung handelt vom DSH-Wrangler, einem interaktiven Werkzeug mit grafischer Benutzeroberfläche (GUI) zur Aufbereitung von tabellarischen, möglicherweise verschachtelten, Daten. Ein langfristiges Ziel des DSH-Wranglers ist die Generierung von DSH-Abfragen entsprechend der Benutzeraktionen in der grafischen Benutzeroberfläche. Die Ausarbeitung beschreibt eine JSON Repräsentation für verschachtelte, tabellarische Daten und ein frühes Stadium der Entwicklung des DSH-Wranglers. Der DSH-Wrangler ist als Webapplikation in JavaScript implementiert und unterstützt die Funktionen und Datentypen von DSH.

Sebastian Engel

Trello Bridge

Use the Trello API to implement services that can be useful to the DB@UTÜ group.

Alexander Ulrich

A Ferry-Based Query Backend for the Links Programming Language

This thesis describes the implementation of a Ferry-based query backend for the Links programming language. In Links, queries are seamlessly embedded into the language: Queries formulated in a subset of the language are translated into single SQL queries. Links uses static checks to ensure that a type-correct query expression can be translated into an equivalent SQL query and allows abstraction over parts of a query. The queryizable subset of Links is, however, severely limited in terms of supported functions and the data type (limited to bags of flat records) of queries.