Real science at the petascale

Abstract

We describe computational science research that uses petascale resources to achieve scientific results at unprecedented scales and resolution. The applications span a wide range of domains, from investigation of fundamental problems in turbulence through computational materials science research to biomedical applications at the forefront of HIV/AIDS research and cerebrovascular haemodynamics. This work was mainly performed on the US TeraGrid 'petascale' resource, Ranger, at Texas Advanced Computing Center, in the first half of 2008 when it was the largest computing system in the world available for open scientific research. We have sought to use this petascale supercomputer optimally across application domains and scales, exploiting the excellent parallel scaling performance found on up to at least 32 768 cores for certain of our codes in the so-called 'capability computing' category as well as highthroughput intermediate-scale jobs for ensemble simulations in the 32-512 core range. Furthermore, this activity provides evidence that conventional parallel programming with MPI should be successful at the petascale in the short to medium term. We also report on the parallel performance of some of our codes on up to 65 636 cores on the IBM Blue Gene/P system at the Argonne Leadership Computing Facility, which has recently been named the fastest supercomputer in the world for open science.

Unstable periodic orbits in the Lorenz attractor

“…Our method is briefly described in Fazendeiro et al [9] and, with much greater detail, in Boghosian et al [10], which should be seen as a companion piece to the present work. This methodology has also been applied to the identification of UPOs in the NSEs [9,11,12], simulated using the lattice Boltzmann method [13], something for which petascale resources are required.…”

Science on the TeraGrid

“…The vasculature is decomposed into a number of computational domains for parallel execution using MPI. Computation is overlapped with communication between different processor domains, resulting in an essentially linear scaling up to 1024 processors and beyond on machines such as Ranger on the TeraGrid [13].…”

“…The vasculature is decomposed into a number of computational domains for parallel execution using MPI. Computation is overlapped with communication between different processor domains, resulting in an essentially linear scaling up to 1024 processors and beyond on machines such as Ranger on the TeraGrid [13].Rather than relying on post-processing visualization, an in-situ volume-rendering approach is used within HemeLB where the ray-traced domains correspond to the LB computational domains. Sub-images are composited to form a complete frame, and transmitted over the network to a lightweight client, resulting in immediate real-time visualization of the bloodflow simulation as the MPI simulation is running.…”

New variational principles for locating periodic orbits of differential equations

“…The space-time variational approach presented in §2 has been applied to the LB equation [11,13,14] force [15][16][17]. For this work, we used a Lattice-Bhatnagar-Gross-Krook (LBGK) model whose main equation can be written as…”