05978096

a

05978096 Albing, Carl Troullier, Norm Whalen, Stephen Olson, Ryan Glenski, Joe Pritchard, Howard Mills, Hugo Scalable node allocation for improved performance in regular and anisotropic 3D torus supercomputers. Cotronis, Yiannis (ed.) et al., Recent advances in the message passing interface. 18th European MPI users' group meeting, EuroMPI 2011, Santorini, Greece, September 18--21, 2011. Proceedings. Berlin: Springer (ISBN 978-3-642-24448-3/pbk). Lecture Notes in Computer Science 6960, 61-70 (2011). 2011 Berlin: Springer EN resource management application placement scheduling topology placement software performance

doi:10.1007/978-3-642-24449-0_9

Summary: MPI application performance can vary based on the scheduler's placing of ranks, whether between nodes or on cores in the same multi-core chip. MPI applications, by default, are at the mercy of the application placement software decision that assigns nodes to a job. We describe herein the general approach of node ordering for allocation in a 3D torus, how it improved MPI application performance, even in the face of an anisotropic interconnect. We demonstrate, quantitatively, that our topologically-based ordering results in improved performance for several MPI applications running on a Top10 supercomputer.