Honing and proofing Astrophysical codes on the road to Exascale. Experiences from code modernization on many-core systems
arXiv:2002.08161 · doi:10.1016/j.future.2020.05.003
Abstract
The complexity of modern and upcoming computing architectures poses severe challenges for code developers and application specialists, and forces them to expose the highest possible degree of parallelism, in order to make the best use of the available hardware. The Intel Xeon Phi of second generation (code-named Knights Landing, henceforth KNL) is the latest many-core system, which implements several interesting hardware features like for example a large number of cores per node (up to 72), the 512 bits-wide vector registers and the high-bandwidth memory. The unique features of KNL make this platform a powerful testbed for modern HPC applications. The performance of codes on KNL is therefore a useful proxy of their readiness for future architectures. In this work we describe the lessons learnt during the optimisation of the widely used codes for computational astrophysics P-Gadget-3, Flash and Echo. Moreover, we present results for the visualisation and analysis tools VisIt and yt. These examples show that modern architectures benefit from code optimisation at different levels, even more than traditional multi-core systems. However, the level of modernisation of typical community codes still needs improvements, for them to fully utilise resources of novel architectures.
arXiv comment16 pages, 10 figures, 4 tables. To be published in Future Generation of Computer Systems (FGCS), Special Issue on "On The Road to Exascale II: Advances in High Performance Computing and Simulations"
References in corpus (23)
- The cosmological simulation code GADGET-2
- A Multi-Code Analysis Toolkit for Astrophysical Simulation Data
- Enzo: An Adaptive Mesh Refinement Code for Astrophysics
- The Star Formation Rate of Turbulent Magnetized Clouds: Comparing Theory, Simulations, and Observations
- Phantom: A smoothed particle hydrodynamics and magnetohydrodynamics code for astrophysics
- On the Star Formation Efficiency of Turbulent Magnetized Clouds
- ECHO: an Eulerian Conservative High Order scheme for general relativistic magnetohydrodynamics and magnetodynamics
- Gasoline2: A Modern SPH Code
- Axisymmetric equilibrium models for magnetized neutron stars in General Relativity under the Conformally Flat Condition
- A fully covariant mean-field dynamo closure for numerical 3+1 resistive GRMHD
- Fast reconnection in relativistic plasmas: the magnetohydrodynamics tearing instability revisited
- Constraints on particle acceleration sites in the Crab Nebula from relativistic MHD simulations
- General relativistic models for rotating magnetized neutron stars in conformally flat spacetime
- Feedback from Reorienting AGN Jets
- SWIFT: Using task-based parallelism, fully asynchronous communication, and graph partition-based domain decomposition for strong scaling on more than 100,000 cores
- Papaloizou-Pringle instability suppression by the magnetorotational instability in relativistic accretion discs
- Adaptive mesh refinement simulations of a galaxy cluster merger - I. Resolving and modelling the turbulent flow in the cluster outskirts
- WOMBAT: A Scalable and High Performance Astrophysical MHD Code
- Optimization and parallelization of B-spline based orbital evaluations in QMC on multi/many-core shared memory processors
- The world's largest turbulence simulations
- Performance Optimisation of Smoothed Particle Hydrodynamics Algorithms for Multi/Many-Core Architectures
- ECHO-3DHPC: relativistic accretion disks onto black holes
- Speeding simulation analysis up with yt and Intel Distribution for Python