基于MPI2的PC集群在离子轨迹仿真中的应用

Qinchun Xie, Yongquan Lu, Yong Liu, Dayu Yin
{"title":"基于MPI2的PC集群在离子轨迹仿真中的应用","authors":"Qinchun Xie, Yongquan Lu, Yong Liu, Dayu Yin","doi":"10.1109/ICISCE.2015.40","DOIUrl":null,"url":null,"abstract":"This paper provides that when the beam current increases to a certain level, the number of ions must be increased and the space charge effect must be added in accelerator physics in order to gain high precise status of ions. It's needed high performance computing to complete this work which needs TFLOPS level capability. So, a PC cluster based on Beowulf architecture and MPI2 was constructed, we put forward a parallel algorithm of divides and conquer according the mapping model of parallel problem, and we accomplished the simulation application of ions trajectory. In the end, we tested the application on different ions' simulation scale from 1.0E+4 to 1.0E+6. Experimental results show that optimized parallel algorithm gets 1.12~2.03 times on average speedup under the premise of guarantee accuracy. The best speedup is 2.52 when the number of ion is 1.0E+6 and the loop times is 1000. The simulation computation time is one third of be-fore. The success transplantation of ion parallel simulation software shows the superiority of the parallel computing and the validity of PC cluster. Additionally, it may be a good reference to other applications.","PeriodicalId":356250,"journal":{"name":"2015 2nd International Conference on Information Science and Control Engineering","volume":"37 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Application of PC Cluster Based on MPI2 in Ions Trajectory Simulation\",\"authors\":\"Qinchun Xie, Yongquan Lu, Yong Liu, Dayu Yin\",\"doi\":\"10.1109/ICISCE.2015.40\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper provides that when the beam current increases to a certain level, the number of ions must be increased and the space charge effect must be added in accelerator physics in order to gain high precise status of ions. It's needed high performance computing to complete this work which needs TFLOPS level capability. So, a PC cluster based on Beowulf architecture and MPI2 was constructed, we put forward a parallel algorithm of divides and conquer according the mapping model of parallel problem, and we accomplished the simulation application of ions trajectory. In the end, we tested the application on different ions' simulation scale from 1.0E+4 to 1.0E+6. Experimental results show that optimized parallel algorithm gets 1.12~2.03 times on average speedup under the premise of guarantee accuracy. The best speedup is 2.52 when the number of ion is 1.0E+6 and the loop times is 1000. The simulation computation time is one third of be-fore. The success transplantation of ion parallel simulation software shows the superiority of the parallel computing and the validity of PC cluster. Additionally, it may be a good reference to other applications.\",\"PeriodicalId\":356250,\"journal\":{\"name\":\"2015 2nd International Conference on Information Science and Control Engineering\",\"volume\":\"37 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2015-04-24\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2015 2nd International Conference on Information Science and Control Engineering\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICISCE.2015.40\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 2nd International Conference on Information Science and Control Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICISCE.2015.40","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

本文提出在加速器物理中,当束流增加到一定程度时,必须增加离子的数量,并加入空间电荷效应,才能获得高精度的离子状态。完成这项工作需要高性能计算,需要TFLOPS级的能力。为此,构建了一个基于Beowulf架构和MPI2的PC集群,根据并行问题的映射模型提出了一种分治并行算法,并完成了离子轨迹的仿真应用。最后,我们在1.0E+4到1.0E+6的不同离子模拟尺度下对应用进行了测试。实验结果表明,优化后的并行算法在保证精度的前提下,平均加速提高了1.12~2.03倍。当离子数为1.0E+6,循环次数为1000次时,最佳加速速度为2.52。仿真计算时间是以前的三分之一。离子并行仿真软件的成功移植表明了并行计算的优越性和PC集群的有效性。此外,它可能是其他应用程序的一个很好的参考。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Application of PC Cluster Based on MPI2 in Ions Trajectory Simulation
This paper provides that when the beam current increases to a certain level, the number of ions must be increased and the space charge effect must be added in accelerator physics in order to gain high precise status of ions. It's needed high performance computing to complete this work which needs TFLOPS level capability. So, a PC cluster based on Beowulf architecture and MPI2 was constructed, we put forward a parallel algorithm of divides and conquer according the mapping model of parallel problem, and we accomplished the simulation application of ions trajectory. In the end, we tested the application on different ions' simulation scale from 1.0E+4 to 1.0E+6. Experimental results show that optimized parallel algorithm gets 1.12~2.03 times on average speedup under the premise of guarantee accuracy. The best speedup is 2.52 when the number of ion is 1.0E+6 and the loop times is 1000. The simulation computation time is one third of be-fore. The success transplantation of ion parallel simulation software shows the superiority of the parallel computing and the validity of PC cluster. Additionally, it may be a good reference to other applications.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信