Institut für Verteilte Systeme

Unser Institut beschäftigt sich mit Themen wie Skalierbarkeit, Zuverlässigkeit, Sicherheit und Datenschutz, Selbstorganisation und Beherrschbarkeit von Komplexität in Verteilten Systemen in einer Vielzahl von Einsatzszenarien wie Cloud-Computing oder Fahrzeug-Fahrzeug-Kommunikation.

In der Lehre decken wir das gesamte Spektrum von Rechnernetzen, über verteilte Systeme bis hin zu Sicherheit und Privacy-Schutz ab.

Unsere letzten Publikationen


Matousek, Matthias; Bösch, Christoph; Kargl, Frank
Poster: Privacy-Preserving Decision Trees
Privacy in Machine Learning and Artificial Intelligence Workshop at ICML 2018
Juli 2018
akzeptiert
Lukaseder, Thomas; Fiedler, Jessika; Kargl, Frank
Performance Evaluation in High-Speed Networks by the Example of Intrusion Detection Systems
11. DFN-Forum Kommunikationstechnologien,
Juni 2018
Matousek, Matthias; Yassin, Mahmoud; Al-Momani, Ala'a; van der Heijden, Rens W.; Kargl, Frank
Robust Detection of Anomalous Driving Behavior
IEEE 87th Vehicular Technology Conference (VTC)
Juni 2018
Mödinger, David; Kopp, Henning; Kargl, Frank; Hauck, Franz J.
Towards Enhanced Network Privacy for Blockchains
Short research statement for the DSN Workshop on Byzantine Consensus and Resilient Blockchains (BCRB)
Juni 2018

Zusammenfassung: Privacy aspects of blockchains have gained attention as the log of transactions can be view by any interested party. Privacy mechanisms applied to the ledger can be undermined by attackers on the network level, resulting in deanonymization of the transaction senders. We discuss current approaches to this problem, e.g. Dandelion, sketch our own approach to provide even stronger privacy mechanisms and discuss the challenges and open questions for further research in this area.

Erb, Benjamin; Meißner, Dominik; Steer, Benjamin A.; Margan, Domagoj; Kargl, Frank; Cuadrado, Felix; Pietzuch, Peter
GraphTides: A Framework for Evaluating Stream-based Graph Processing Platforms
Proceedings of the 1st Joint International Workshop on Graph Data Management Experiences & Systems (GRADES) and Network Data Analytics (NDA)
Juni 2018
akzeptiert

Zusammenfassung: Stream-based graph systems continuously ingest graph-changing events via an established input stream, performing the required computation on the corresponding graph. While there are various benchmarking and evaluation approaches for traditional, batch-oriented graph processing systems, there are no common procedures for evaluating stream-based graph systems. We, therefore, present GraphTides, a generic framework which includes the definition of an appropriate system model, an exploration of the parameter space, suitable workloads, and computations required for evaluating such systems. Furthermore, we propose a methodology and provide an architecture for running experimental evaluations. With our framework, we hope to systematically support system development, performance measurements, engineering, and comparisons of stream-based graph systems.

Export als: BibTeX, XML

Klicken Sie hier um eine Übersicht aller Publikationen zu erhalten.