Optimizing the hybrid parallelization of BHAC

نویسندگان

چکیده

We present our experience with the modernization on GR-MHD code BHAC, aimed at improving its novel hybrid (MPI+OpenMP) parallelization scheme. In doing so, we showcase use of performance profiling tools usable x86 (Intel-based) architectures. Our characterization and threading analysis provided guidance in concurrency thus efficiency OpenMP parallel regions. assess scaling communication patterns order to identify alleviate MPI bottlenecks, both runtime switches precise interventions. The optimized version BHAC improved by ∼28%, making it viable for several hundreds supercomputer nodes. finally test whether porting such optimizations different hardware is likewise beneficial new architecture running ARM A64FX vector

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Optimizing HPC and Parallelization For Nanotechnology

This paper describes how high performance computing can be applied for computations in Nanotechnology. We have discussed the problems that exist in a High Performance Computing (HPC) arrangement. We have put forth a proposition to optimize the computations in an HPC setup. The distribution of work in various clusters has been formulated to make Nanotechnology computations more effective and rea...

متن کامل

Optimizing SIEM Throughput on the Cloud Using Parallelization

Processing large amounts of data in real time for identifying security issues pose several performance challenges, especially when hardware infrastructure is limited. Managed Security Service Providers (MSSP), mostly hosting their applications on the Cloud, receive events at a very high rate that varies from a few hundred to a couple of thousand events per second (EPS). It is critical to proces...

متن کامل

Correction: Optimizing SIEM Throughput on the Cloud Using Parallelization

[This corrects the article DOI: 10.1371/journal.pone.0162746.].

متن کامل

Hybrid MPI-OpenMP Parallelization of Image Reconstruction

Perceptual coding should take full advantage of the results from human visual system (HVS) studies. In this paper, we exploit an image reconstruction algorithm, which can simply and reliably represent images using regional shape and texture information and effectually eliminate perceptual redundancy. But real-time application of the algorithm is limited due to its inherent complex and intensive...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Astronomy and Computing

سال: 2022

ISSN: ['2213-1345', '2213-1337']

DOI: https://doi.org/10.1016/j.ascom.2021.100509