Dear all,

 

I have an open PhD position in Network Algorithms at the University of Vienna, see below and at:

https://tinyurl.com/y9plkw5b

for more details.

 

The specific topic will depend on the interests and skills of the student, and could revolve around e.g., self-adjusting networks, online graph partitioning, graph embedding (“virtual network embedding”), (semi-)oblivious routing, routing table compression, or self-stabilizing algorithms. Similarly, the methodology will depend on the student and includes graph algorithms, online algorithms, approximation algorithms, learning algorithms, linear programming and randomized rounding, as well as computer-generated algorithms. See below some example papers.

 

For an overview of our research group as well as publications, see https://ct.cs.univie.ac.at/ and https://net.t-labs.tu-berlin.de/~stefan/ .

 

Please forward to interested students and do not hesitate to contact me for more information!

 

Many thanks and kind regards,

Stefan

 

Synopsis:

 

In contrast to traditional communication networks which rely on proprietary algorithms and “blackbox” hardware, modern networks support numerous algorithmic optimizations: in terms of routing, traffic engineering, medium access, queuing, etc. become „programmable“ and “software-defined”. Companies like Google exploit these optimization opportunities, e.g., to efficiently schedule big data transfers across their wide-area network. Many companies also exploit these optimization opportunities to allocate flows in a datacenter network more efficiently and, e.g., reduce latency or job completion time. As datacenter networks are typically also highly virtualized, additional optimization opportunities arise: for example it is possible to flexibly place or even migrate communication endpoints (virtual machines or containers), to reduce communication costs (keywords: „resource allocation“ and „virtual network embedding“). Recently, it has even become possible to change the physical network topology over time, enabling “demand-aware networks” whose topology is optimized toward the traffic pattern or workload they serve, and introducing novel „network design“ problems (keyword: self-adjusting networks). 

Yet, the algorithmic problems underlying modern communication networks, while practically very relevant, are not well understood today. We hence are looking for a PhD student (“pre-doc”/”Doktorand(in)”) with a strong background and interest in algorithms (in particular, graph algorithms, approximation algorithms, randomized algorithms, distributed algorithms, dynamic programming, linear programming etc.). The goal of our research is to lay the algorithmic foundations of modern communication networks, e.g., in areas such as routing, scheduling, clustering, and network design. It is expected that the PhD student contributes to the modelling, algorithm design, optimization, and formal analysis (e.g., proofs of correctness and performance) for such networks.

 

Some related publications:

 

SplayNet: Towards Locally Self-Adjusting Networks
Stefan Schmid, Chen Avin, Christian Scheideler, Michael Borokhovich, Bernhard Haeupler, and Zvi Lotker.
IEEE/ACM Transactions on Networking (TON), Volume 24, Issue 3, 2016.
Documents: paper 
pdf, bibtex bib

 

Online Balanced Repartitioning
Chen Avin, Andreas Loukas, Maciej Pacut, and Stefan Schmid.
30th International Symposium on Distributed Computing (DISC), Paris, France, September 2016.
Documents: paper pdf, slides pdf, bibtex bib

 

Demand-Aware Network Designs of Bounded Degree
Chen Avin, Kaushik Mondal, and Stefan Schmid.
31st International Symposium on Distributed Computing (DISC), Vienna, Austria, October 2017.
Documents: paper pdf, slides pdf, bibtex bib

 

Congestion-Free Rerouting of Flows on DAGs
Saeed Akhoondian Amiri, Szymon Dudycz, Stefan Schmid, and Sebastian Wiederrecht.
45th International Colloquium on Automata, Languages, and Programming (ICALP), Prague, Czech Republic, July 2018.
Documents: paper pdf, bibtex bib

 

 

Link to apply: https://univis.univie.ac.at/ausschreibungstellensuche/flow/bew_ausschreibung-flow?_flowExecutionKey=_cE64C9E85-DB7C-FACB-B491-830B066F39DF_k8A2068DA-0836-BB8F-4BAD-58B13A83CE1B&tid=70132.28

 

 

--

 

Stefan Schmid

Professor, Computer Science

University of Vienna, Austria

http://cs.univie.ac.at/ct/news/

https://net.t-labs.tu-berlin.de/~stefan/