|
|||||
|
|
| 高校科研团队Microwulf并行计算系统的搭建 | |
| 引用本文： | 刘新荣,胡元,罗建华,葛华.高校科研团队Microwulf并行计算系统的搭建[J].高等建筑教育,2011,20(2):137-141. |
| 作者姓名： | 刘新荣 胡元 罗建华 葛华 |
| 作者单位： | 1. 重庆大学土木工程学院,重庆,400044;后勤工程学院军事建筑工程系,重庆,401311 2. 重庆大学土木工程学院,重庆,400044 3. 成都地质调查中心,四川成都,610081 |
| 基金项目： | 国家自然科学基金创新群体基金，中国地质调查局项目"地震滑坡灾害编图方法示范研究" |
| 摘 要： | 由于成本限制,高校科研团队的计算机多为普通型个人电脑(PC),无法执行涉及多节点的并行计算任务。基于Beowulf架构与可于大众电脑市场采购到的普通硬件设备搭建了一套四节点8CPU核Microwulf并行运算系统(命名为LXR01)。该系统各节点均安装Windows7与Linux操作系统,其中服务器节点利用NFS建立共享文件系统,分别以gcc、OpenMPI为主要编译环境和并行计算环境。HPL测试表明,LXR01系统的最大运算能力为39.93Gflops,其计算效率和成本效率分别为81.6%和Y-583.8/Gflops。同时,SPECFEM3D代码的计算实例证明LXR01系统能有效地解决并行计算问题。 |
| 关 键 词： | 并行计算 Microwulf Linux OpenMPI |
Microwulf parallelism system in university research team |
|
| LIU Xin-rong,HU Yuan-xin,LUO Jian-hua and GE Hua.Microwulf parallelism system in university research team[J].Journal of Architectural Education in Institutions of Higher Learning,2011,20(2):137-141. | |
| Authors: | LIU Xin-rong HU Yuan-xin LUO Jian-hua and GE Hua |
| Affiliation: | 1.College of Civil Engineering,Chongqing University,Chongqing 400045,P.R.China; 2.Department of Architecture and Civil Engineering,Logistical Engineering University,Chongqing 401311,P.R.China; 3.Nanjiang Hydrogeology and Engineering Geology Brigade,Chongqing 401121,P.R.China; 4.Chengdu Center of China Geological Survey,Chengdu 610081,P.R.China) |
| Abstract: | On account of cost restriction,the computers of university research team are almost personal computers,which can not perform parallel computational tasks related to multiple computer nodes.Based on architecture of Beowulf and common hardware components bought from mass PC market,a Microwulf parallelism system with four nodes and eight CPU cores,named LXR01,was built.Windows7 and Linux operating system were installed on each node of the LXR01 system.The master node of LXR01 built shared file system with NFS,and adopted gcc and OpenMPI as main compiler and parallelism environment,respectively.The LXR01 system was benchmarked using HPL,from which 39.93 Gflops was max performance and the computational efficiency and cost efficiency of LXR01 system was 81.6% and Y-583.8/Gflops respectively.Meanwhile,it was concluded from parallel computational example of SPECFEM3D codes on LXR01 system that the LXR01 system can solve parallel computational problems effectively. |
| Keywords: | parallel computation Microwulf Linux OpenMPI |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |
| 点击此处可从《高等建筑教育》浏览原始摘要信息 | |
| 点击此处可从《高等建筑教育》下载全文 | |