TUD Logo

TUD Startseite » ... » Fakultät Informatik » Forschung » Wissenschaftliche Vorträge

Fakultät Informatik

against racism

Wissenschaftliche Vorträge

Routing on the Channel Dependency Graph: A New Approach to Deadlock-Free, Destination-Based, High-Performance Routing for Lossless Interconnection Networks

Vortrag im Promotionsverfahren (Informatik) von Jens Domke

7.12.2015, 16:00 Uhr, APB 1096

Lossless interconnection networks become more popular in both, high performance and data center computing. Such networks require deadlock-free routing algorithms to utilize the available hardware efficiently. However, failures in hardware and software are omnipresent and often inevitable, which results in a decrease of the mean-time-between-failure for growing system sizes. The performance of inter- and intra-program communication on those systems is influenced by the interconnection network, as well as the routing algorithm in use.

Topology-aware routing algorithms become increasingly inapplicable, due to irregular topologies, which either are irregular by design or as a result of hardware failures. Existing topology-agnostic routing methods either suffer from poor path balancing or are not bounded in the number of so-called "virtual channels" which are needed to resolve deadlocks in the routing tables. In this presentation, I will survey prevalent and recently proposed oblivious routing algorithms from both categories. Additionally, I will review the basics of deadlocks or credit loops, as well as techniques to avoid these. As a result, I will deduce the necessity for a topology-agnostic, general-purpose routing algorithm, which enforces deadlock-freedom while optimizing the path balancing within given constraints, and is applicable to state-of-the-art and future interconnection technology.

Diese Veranstaltung wird unterstützt von ZIH.

Query-Time Data Integration

Verteidigung im Promotionsverfahren von Dipl.-Medieninf. Julian Eberius (Institut für Systemarchitektur, Professur Datenbanken)

10.12.2015, 14:00 Uhr, APB 3105 (Beratungsraum 3. Etage

Today, data is collected in ever increasing scale and variety, opening up enormous potential for new insights and data-centric products. However, in many cases the volume and heterogeneity of new data sources precludes up-front integration using traditional ETL processes and data warehouses. In some cases, it is even unclear if and in what context the collected data will be utilized. Therefore, there is a need for agile methods that defer the effort of integration until the usage context is established. This thesis introduces Query-Time Data Integration as an alternative concept to traditional up-front integration. It aims at enabling users to issue ad-hoc queries on their own data as if all potential other data sources were already integrated, without declaring specific sources and mappings to use. Automated data search and integration methods are then coupled directly with query processing on the available data. The ambiguity and uncertainty introduced through fully automated retrieval and mapping methods is compensated by answering those queries with ranked lists of alternative results. Each result is then based on different data sources or query interpretations, allowing users to pick the result most suitable to their information need. To this end, this thesis makes three main contributions. Firstly, we introduce a novel method for Top-k Entity Augmentation, which is able to construct a top-k list of consistent integration results from a large corpus of heterogeneous data sources. It improves on the state-of-the-art by producing a set of individually consistent, but mutually diverse, set of alternative solutions, while minimizing the number of data sources used. Secondly, based on this novel augmentation method, we introduce the DrillBeyond system, which is able to process Open World SQL queries, i.e., queries referencing arbitrary attributes not defined in the queried database. The original database is then augmented at query time with Web data sources providing those attributes. Its hybrid augmentation/relational query processing enables the use of ad-hoc data search and integration in data analysis queries, and improves both performance and quality when compared to using separate systems for the two tasks. Finally, we studied the management of large-scale dataset corpora such as data lakes or Open Data platforms, which are used as data sources for our augmentation methods. We introduce Publish-time Data Integration as a new technique for data curation systems managing such corpora, which aims at improving the individual reusability of datasets without requiring up-front global integration. This is achieved by automatically generating metadata and format recommendations, allowing publishers to enhance their datasets with minimal effort. Collectively, these three contributions are the foundation of a Query-time Data Integration architecture, that enables ad-hoc data search and integration queries over large heterogeneous dataset collections.

Flexible Aspect-Based Encoding for Resilient Applications

Präsentation der Bachelor-Arbeit (Informatik) von Bàlint Gyapjas (Institut für Technische Informatik, Professur Compilerbau)

10.12.2015, 14:50 Uhr, APB 2101 Beratungsraum 2. Etage

In safety critical areas it is important that results of computations are reliable. One way to achieve this is through the encoding of programs, which is an approach to add error detection capabilities to software. Encoding modifies the values that a program processes but also the program's instructions themselves. The automatization of this task has been the subject of past research. In this work we investigate an aspect-oriented approach to implement program encoding. Aspect-Oriented Programming is a method from software engineering that allows the encapsulation of systemic program properties, known as cross-cutting concerns. Using aspect-orientation to separate the encoding implementation from the rest of the program enables its reusability to arbitrary programs.

Diese Veranstaltung wird unterstützt von Professur Compilerbau.

Suche im Ankündigungsarchiv

Abonnieren Sie die Vortragsankündigungen als News Feed: RSS

Stand: 1.12.2015, 17:26 Uhr
Autor: Webmaster