Optimizing Linpack Benchmark on GPU-Accelerated Petascale Supercomputer.
Saved in:
| Title: | Optimizing Linpack Benchmark on GPU-Accelerated Petascale Supercomputer. |
|---|---|
| Authors: | Wang, Feng1 fengwang@nudt.edu.cn, Yang, Can-Qun1 canqun@nudt.edu.cn, Du, Yun-Fei1 duyunfei@nudt.edu.cn, Chen, Juan1 juanchen@nudt.edu.cn, Yi, Hui-Zhan1 huizhanyi@nudt.edu.cn, Xu, Wei-Xia1 xuwx@nudt.edu.cn |
| Source: | Journal of Computer Science & Technology (10009000). Sep2011, Vol. 26 Issue 5, p854-865. 12p. |
| Subjects: | LINPACK (Computer system), Benchmarking (Management), Graphics processing units, Supercomputers, Program transformation, Heterogeneous computing |
| Geographic Terms: | China |
| Abstract: | In this paper we present the programming of the Linpack benchmark on TianHe-1 system, the first petascale supercomputer system of China, and the largest GPU-accelerated heterogeneous system ever attempted before. A hybrid programming model consisting of MPI, OpenMP and streaming computing is described to explore the task parallel, thread parallel and data parallel of the Linpack. We explain how we optimized the load distribution across the CPUs and GPUs using the two-level adaptive method and describe the implementation in details. To overcome the low-bandwidth between the CPU and GPU communication, we present a software pipelining technique to hide the communication overhead. Combined with other traditional optimizations, the Linpack we developed achieved 196 :7 GFLOPS on a single compute element of TianHe-1. This result is 70 :1% of the peak compute capability, 3 :3 times faster than the result by using the vendor's library. On the full configuration of TianHe-1 our optimizations resulted in a Linpack performance of 0 :563 PFLOPS, which made TianHe-1 the 5th fastest supercomputer on the Top500 list in November, 2009. [ABSTRACT FROM AUTHOR] |
| Copyright of Journal of Computer Science & Technology (10009000) is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
| FullText | Text: Availability: 0 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 65796946 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Optimizing Linpack Benchmark on GPU-Accelerated Petascale Supercomputer. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Wang%2C+Feng%22">Wang, Feng</searchLink><relatesTo>1</relatesTo><i> fengwang@nudt.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Yang%2C+Can-Qun%22">Yang, Can-Qun</searchLink><relatesTo>1</relatesTo><i> canqun@nudt.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Du%2C+Yun-Fei%22">Du, Yun-Fei</searchLink><relatesTo>1</relatesTo><i> duyunfei@nudt.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Chen%2C+Juan%22">Chen, Juan</searchLink><relatesTo>1</relatesTo><i> juanchen@nudt.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Yi%2C+Hui-Zhan%22">Yi, Hui-Zhan</searchLink><relatesTo>1</relatesTo><i> huizhanyi@nudt.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Xu%2C+Wei-Xia%22">Xu, Wei-Xia</searchLink><relatesTo>1</relatesTo><i> xuwx@nudt.edu.cn</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Journal+of+Computer+Science+%26+Technology+%2810009000%29%22">Journal of Computer Science & Technology (10009000)</searchLink>. Sep2011, Vol. 26 Issue 5, p854-865. 12p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22LINPACK+%28Computer+system%29%22">LINPACK (Computer system)</searchLink><br /><searchLink fieldCode="DE" term="%22Benchmarking+%28Management%29%22">Benchmarking (Management)</searchLink><br /><searchLink fieldCode="DE" term="%22Graphics+processing+units%22">Graphics processing units</searchLink><br /><searchLink fieldCode="DE" term="%22Supercomputers%22">Supercomputers</searchLink><br /><searchLink fieldCode="DE" term="%22Program+transformation%22">Program transformation</searchLink><br /><searchLink fieldCode="DE" term="%22Heterogeneous+computing%22">Heterogeneous computing</searchLink> – Name: SubjectGeographic Label: Geographic Terms Group: Su Data: <searchLink fieldCode="DE" term="%22China%22">China</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: In this paper we present the programming of the Linpack benchmark on TianHe-1 system, the first petascale supercomputer system of China, and the largest GPU-accelerated heterogeneous system ever attempted before. A hybrid programming model consisting of MPI, OpenMP and streaming computing is described to explore the task parallel, thread parallel and data parallel of the Linpack. We explain how we optimized the load distribution across the CPUs and GPUs using the two-level adaptive method and describe the implementation in details. To overcome the low-bandwidth between the CPU and GPU communication, we present a software pipelining technique to hide the communication overhead. Combined with other traditional optimizations, the Linpack we developed achieved 196 :7 GFLOPS on a single compute element of TianHe-1. This result is 70 :1% of the peak compute capability, 3 :3 times faster than the result by using the vendor's library. On the full configuration of TianHe-1 our optimizations resulted in a Linpack performance of 0 :563 PFLOPS, which made TianHe-1 the 5th fastest supercomputer on the Top500 list in November, 2009. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of Journal of Computer Science & Technology (10009000) is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=65796946 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1007/s11390-011-0184-1 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 12 StartPage: 854 Subjects: – SubjectFull: LINPACK (Computer system) Type: general – SubjectFull: Benchmarking (Management) Type: general – SubjectFull: Graphics processing units Type: general – SubjectFull: Supercomputers Type: general – SubjectFull: Program transformation Type: general – SubjectFull: Heterogeneous computing Type: general – SubjectFull: China Type: general Titles: – TitleFull: Optimizing Linpack Benchmark on GPU-Accelerated Petascale Supercomputer. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Wang, Feng – PersonEntity: Name: NameFull: Yang, Can-Qun – PersonEntity: Name: NameFull: Du, Yun-Fei – PersonEntity: Name: NameFull: Chen, Juan – PersonEntity: Name: NameFull: Yi, Hui-Zhan – PersonEntity: Name: NameFull: Xu, Wei-Xia IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 09 Text: Sep2011 Type: published Y: 2011 Identifiers: – Type: issn-print Value: 10009000 Numbering: – Type: volume Value: 26 – Type: issue Value: 5 Titles: – TitleFull: Journal of Computer Science & Technology (10009000) Type: main |
| ResultId | 1 |
