fluent6.2双核并行效率（FLUENT6.2 dual core parallel efficiency）.docVIP

下载本文档

5
0
约4.34千字
约 5页
2017-08-31 发布于河南
举报
版权申诉

fluent6.2双核并行效率（FLUENT6.2 dual core parallel efficiency）.doc

1、本文档共5页，可阅读全部内容。
2、有哪些信誉好的足球投注网站（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。
3、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
5、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
6、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
7、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
8、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

fluent6.2双核并行效率（FLUENT6.2 dual core parallel efficiency）

fluent6.2双核并行效率（FLUENT6.2 dual core parallel efficiency） FLuent6.2 dual core parallel efficiency Until the use of Athlon X2 Fluent dual core parallel calculation, the efficiency is very high, the speed is almost 2 times of the single core, so in front of Conroe test platform, no single core comparison test, because it is assumed that the subjective is approximately 2 times the performance of dual core single core, but recently inadvertently ran a E6300 the single kernel calculation, found its not so, so today made a detailed comparison test. The hardware platform, test software (Fluent 6.2) and the test case with a Fluent 6.2 benchmark:Conroe VS Athlon64 test X2 exactly the same, only the system platform for some changes, windows or XP-64 system, Linux system, E6300 platform has been replaced recently released RHEL5 (Redhat Enterprise Linux Server x86_64 (beta2) kernel-2.6.18-1.2747.el5), X2 3800+ platform Linux replaced the earlier SUSE 10 x86_64 (kernel-smp-2.6.13-15), according to the previous test and experience, X2 3800+ if it replaced RHEL5 beta2, should be a little bit faster calculation, for this example for each time step to speed up the time probably within 1s, relatively speaking can be neglected. When testing dual core parallel, the Fluent smpi Memory (Shared Port Interface) is adopted, which is more suitable for single CPU or multi-core platform. The test results are as follows: (Statistics average time consuming per step, unit s (SEC)): CPU frequency winXP-64, Linux, X86_64 Mononuclear double core Conroe, E6300, 400*7=2.8G, 39.498, 22.703, 28.756, 17.507 329*7=2.3G 47.873, 28.487, 34.960, 21.177 AMD, X2, 3800+, 230*10=2.3G, 76.603, 39.602, 46.198, 23.374 Conclusion: Single core performance of Conroe is very good, the same frequency (2.3G), Win32 60% faster than the X2, Linux x86_64 32% faster than the X2, but not as dual parallel efficiency, only 1.68-1.74 times of single core performance achieved in Win32, only 1.64-1.65 times of single core performan