Large-Scale Network Simulation
Time 01/16/13 03:30PM-05:00PM
Looking at the thousands of biologists struggling to understand basic phenomena such as functional/structural dependence in proteins, it may seem that there is little chance of ever modeling the Internet at large scale. However, unlike natural phenomena, the Internet is engineered by people, and thus is fundamentally better understood at small-scale, as well as fundamentally more simple than systems resulting from billions of years of incremental change. Thus, despite the absence of global reductionist principles, a mechanistic approach to modeling large networks can build on our understanding of the engineering of individual components, which still are relatively small in number compared to biological systems. To achieve this goal, we need to harness the massive computational power of parallel computers, to extrapolate from the components we understand to the behaviors we can observe, through simulation of those components. This session will discuss approaches to network simulation at large scale, such as discrete event simulation, flow graphs, and model checking.
Speaker Taghrid Samak Lawrence Berkeley National Laboratory
Speaker Martin Swany Indiana University