Tag Archives: Programming Languages

GPU Acceleration of an Established Solar MHD Code using OpenACC [CL]

Posted on by

http://arxiv.org/abs/1811.02605


GPU accelerators have had a notable impact on high-performance computing across many disciplines. They provide high performance with low cost/power, and therefore have become a primary compute resource on many of the largest supercomputers. Here, we implement multi-GPU acceleration into our Solar MHD code (MAS) using OpenACC in a fully portable, single-source manner. Our preliminary implementation is focused on MAS running in a reduced physics “zero-beta” mode. While valuable on its own, our main goal is to pave the way for a full physics, thermodynamic MHD implementation. We describe the OpenACC implementation methodology and challenges. “Time-to-solution” performance results of a production-level flux rope eruption simulation on multi-CPU and multi-GPU systems are shown. We find that the GPU-accelerated MAS code has the ability to run “zero-beta” simulations on a single multi-GPU server at speeds previously requiring multiple CPU server-nodes of a supercomputer.

Read this paper on arXiv…

R. Caplan, J. Linker, Z. Mikić, et. al.
Thu, 8 Nov 18
40/72

Comments: 13 pages, 9 figures

Studying tidal effects in planetary systems with Posidonius. A N-body simulator written in Rust [EPA]

Posted on by

http://arxiv.org/abs/1712.01281


Planetary systems with several planets in compact orbital configurations such as TRAPPIST-1 are surely affected by tidal effects. Its study provides us with important insight about its evolution. We developed a second generation of a N-body code based on the tidal model used in Mercury-T, re-implementing and improving its functionalities using Rust as programming language (including a Python interface for easy use) and the WHFAST integrator. The new open source code ensures memory safety, reproducibility of numerical N-body experiments, it improves the spin integration compared to Mercury-T and allows to take into account a new prescription for the dissipation of tidal inertial waves in the convective envelope of stars. Posidonius is also suitable for binary system simulations with evolving stars.

Read this paper on arXiv…

S. Blanco-Cuaresma and E. Bolmont
Wed, 6 Dec 17
68/71

Comments: To appear in the “EWASS Special Session 4 (2017): Star-planet interactions” proceedings

What can the programming language Rust do for astrophysics? [IMA]

Posted on by

http://arxiv.org/abs/1702.02951


The astrophysics community uses different tools for computational tasks such as complex systems simulations, radiative transfer calculations or big data. Programming languages like Fortran, C or C++ are commonly present in these tools and, generally, the language choice was made based on the need for performance. However, this comes at a cost: safety. For instance, a common source of error is the access to invalid memory regions, which produces random execution behaviors and affects the scientific interpretation of the results.
In 2015, Mozilla Research released the first stable version of a new programming language named Rust. Many features make this new language attractive for the scientific community, it is open source and it guarantees memory safety while offering zero-cost abstraction.
We explore the advantages and drawbacks of Rust for astrophysics by re-implementing the fundamental parts of Mercury-T, a Fortran code that simulates the dynamical and tidal evolution of multi-planet systems.

Read this paper on arXiv…

S. Blanco-Cuaresma and E. Bolmont
Mon, 13 Feb 17
26/57

Comments: To appear in the proceedings of the IAU Symposium 325 on Astroinformatics