文章基本信息

标题：A SCABILITY STUDY OF SGI CLUSTERED XFS USING HDF BASED AMR APPLICATION
本地全文：下载
作者：S. Saini ; D. Talcott ; H. Yeung 等
期刊名称：Computational Methods in Science and Technology
印刷版ISSN：1505-0602
出版年度：2006
卷号：12
期号：Special issue
页码：47-54
出版社：Poznan Supercomputing and Networking Center
摘要：Many large-scale parallel scientific and engineering applications, especially climate modeling, often run for lengthy periodsand require data checkpointing periodically to save the state of the computation for a program restart. In addition, such applications needto write data to disks for post-processing, e.g., visualization. Both these scenarios involve a write-only pattern using Hierarchal Data Format (HDF) files. In this paper, we study the scalability of CXFS by HDF based Structured Adaptive Mesh Refinement (AMR) appli-cation for three different block sizes. The code used is a block-structured AMR hydrodynamics code that solves compressible, reactivehydrodynamic equations and characterizes physics and mathematical algorithms used in studying nuclear flashes on neutron stars andwhite dwarfs. The computational domain is divided into blocks distributed across the processors. Typically, a block contains 8 zones ineach coordinate direction (x, y, and z) and a perimeter of guard cells (in this case, 4 zones deep) to hold information from the neighbors.We used three different block sizes of 8 × 8 × 8, 16 × 16 × 16, and 32 × 32 × 32. Results of parallel I/O bandwidths (checkpoint file andtwo plot files) are presented for all three-block sizes on a wide range of processor counts, ranging from 1 to 508 processors of the Co-lumbia system
关键词：parallel I/O; clustered file system (CXFS); benchmarking; performance evaluation; HDF5; adaptive mesh refinement; AMR