Mapping to Irregular Torus Topologies and Other Techniques for Petascale Biomolecular Simulation

Currently deployed petascale supercomputers typically use toroidal network topologies in three to five dimensions. While these networks perform well for topology-agnostic codes on a few thousand nodes, leadership machines with 20,000 nodes require topology awareness to avoid network contention for communication-intensive codes. Topology adaptation is complicated by irregular node allocation shapes and holes due to dedicated input/output nodes or hardware failure. In the context of the popular molecular dynamics program NAMD, we present methods for mapping a periodic 3-D grid of fixed-size spatial decomposition domains to 3-D Cray Gemini and 5-D IBM Blue Gene/Q toroidal networks to enable hundred-million atom full machine simulations, and to similarly partition node allocations into compact domains for smaller simulations using multiple-copy algorithms. Additional enabling techniques are discussed and performance is reported for NCSA Blue Waters, ORNL Titan, and ANL Mira.

@inproceedings{namdSC14,
  author = "James C. Phillips and Yanhua Sun and Nikhil Jain and Eric J. Bohm and Laxmikant V. Kale",
  title = "{Mapping to Irregular Torus Topologies and Other Techniques for Petascale Biomolecular Simulation}",
  booktitle = "Proceedings of ACM/IEEE SC 2014",
  year = "2014",
  month = "November",
  address = {New Orleans, Louisiana},
}