Towards Billion-Scale Social Simulation

Invited talk at SSC2016 by Prof. Toyotaro Suzumura, IBM T.J. Watson Research Center


Many social simulations can be represented using mobile-agent-based model in which agents moving around on a given space such a evacuations, traffic flow and epidemics. Whole planet simulation with
billions of agents at microscopic level helps mitigate the global crisis. It introduces new technical challenges such as processing and migrating many agents and load balancing among hundreds of machines.
To overcome these challenges, well-designed software architecture of a simulator is essential. In this research, we proposed agent-based complex cellular automata architecture (ABCCA) and studied the performance and scalability of two cell-based processing models,through simple traffic flow simulation on multi-core distributed system. The experiments show that the computation speedup can be achieved by reducing granularity of tasks and processing only active spaces. We achieved running the traffic flow simulation with one billion of agents in almost real time on 1,536 CPU cores of total 128 machines of TSUBAME supercomputer

Short Bio:

Prof. Toyotaro Suzumura is currently a research staff member at IBM T.J. Watson Research Center, the headquarter of IBM Research and working on high performance graph analytics platform and social simulation. He is also visiting full professor at Barcelona Supercomputing Center.

His current research topics include big data processing middlewares, large-scale graph analytics, supercomputing / high performance computing, and microscopic simulation platforms. He is now a co-principal investigator for two Japanese government projects funded by Japan Science Foundation that aim at building next-generation big data processing middlewares and libraries. One of his notable achievements is that he and his team won the first place at the world competition of “Big Data” processing on supercomputers called “Graph500” in June 2014, June and November in 2015 by proposing a novel and highly scalable graph search algorithm on massively parallel and distributed computers and successfully implementing it on the Japanese K supercomputer. He has also led a project “ScaleGraph” of developing an open source library for billion-scale graph analytics.

← Older Newer →