Yannic Maus

"Distributed Algorithms for Graph Problems and Beyond"

With the fast growth of networks and the rapid increase of the size of our data there is an ever-increasing need to understand distributed systems and decentralized data processing. In this talk, I will discuss recent developments in the area of distributed graph algorithms/network algorithms, with a focus on classic message passing algorithms. In this setting, we have a network of computers that communicate via fixed communication links. The main challenge in solving graph problems is that the nodes have to base their decision on only local information about the huge network.
I will begin with an introduction to the area through the lens of classic graph coloring problems. Then, I will survey new techniques that lead to efficient distributed algorithms for many classic problems, including vertex

Bio: and edge coloring problems but also optimization problems. While these algorithms are very efficient, they rely on communication links with a large bandwidth which is often infeasible in practice. Thus, I will explain how studying stronger limits on the communication in distributed systems will play a major role in the future of the area and beyond.

Termin als Datei downloaden und in den Kalender importieren.