Axhausen, Kay W.
- Working Paper
Today, agent based micro-simulations are widely used in the field of transport planning and 7 traffic management. One important requirement is the ability to simulate large scale scenarios 8 in reasonable time. An obvious approach to reduce the computation time of such scenarios is 9 to use multiple CPU cores.</br> 10 This paper presents the implementation of a parallel queue simulation for MATSim written 11 in Java. Existing parallel traffic micro-simulations are reviewed concerning their paralleliza12 tion approaches as well as the reached performance gains. Various concepts how to model 13 the progress of time and how to distribute computational workload among multiple CPU cores 14 are discussed. Based on an analysis of the MATSim framework regarding its structure, per15 formance and extensibility the concepts for the parallel queue simulation are selected and im16 plemented. Performance tests with different sized scenarios are conducted. An analysis of the 17 results shows that especially for large scale scenarios a significant performance gain is reach18 able Show more
External linksSearch via SFX
Journal / seriesArbeitsberichte Verkehrs- und Raumplanung
PublisherIVT, ETH Zürich
Organisational unit03521 - Axhausen, Kay W.
02226 - NSL - Netzwerk Stadt und Landschaft / NSL - Network City and Landscape
MoreShow all metadata