AG 1, AG 2, AG 3, AG 4   Info

AG Audience

English

45 Minutes

Saarbrücken

Distributed computing is about loosely coupled machines communicating over heterogeneous networks via channels of various characteristics. The participating machines sometimes crash, and even if they do not then the software they run is often not fully reliable. This poses a challenge to design robust algorithms that can achieve the computational goals in such environments in any contingency. This is even more challenging if we seek efficient solutions, in terms of time or the amount of communication.

We will focus on selected algorithmic problems, such as consensus, gossiping, do-all, and discuss recent work in these areas. Among the issues covered, we will show that randomization sometimes helps. Finally we will survey the remaining open problems.

Spyros Kontogiannis

--email hidden

passcode not visible

logged in users only