Title: Distributed computing on the (fruit) fly (slides in PDF)
Yuval Emek (ETH Zurich, Switzerland)
http://www.disco.ethz.ch/members/yuval.html
Abstract: The study of distributed network algorithms is strongly motivated by the thriving Internet age. As such, the traditional message passing models focus mainly on computer-like devices that can exchange large messages with their network neighbors and perform arbitrary local computations. Recently, there is a trend to apply distributed computing methods and techniques to the study of sub-microprocessor networks, specifically, biological cellular networks. However, the suitability of the traditional message passing models to this type of networks is questionable: do tiny cells "compute" and/or "communicate" essentially the same as a computer?
In this talk we discuss several alternatives to the traditional message passing models that better capture biological cellular networks. In particular, we introduce a new model that depicts a network of finite state machines operating in an asynchronous environment. Although the computation and communication capabilities of each individual device in the new model are, by design, much weaker than those of a computer, we show that some of the most important and extensively studied distributed computing problems can still be solved efficiently.The talk will be self-contained.
