Compiler-Directed Energy Optimization for Parallel-Disk-Based Systems.
Saved in:
| Title: | Compiler-Directed Energy Optimization for Parallel-Disk-Based Systems. |
|---|---|
| Authors: | Seung Woo Son1 sson@cse.psu.edu, Guangyu Chen2 ozturk@cse.psu.edu, Ozturk, Ozcan1 kandemirj@cse.psu.edu, Kandemir, Mahmut1 guchen@microsoft.com, Choudhary, Alok3 choudhar@ece.northwestern.edu |
| Source: | IEEE Transactions on Parallel & Distributed Systems. Sep2007, Vol. 18 Issue 9, p1241-1257. 17p. 9 Diagrams, 4 Charts, 11 Graphs. |
| Subjects: | Optimizing compilers, Compilers (Computer programs), Algorithms, Parallel algorithms, Linear programming, Electronic file management, Computer programming, Computer storage devices, Computer industry |
| Abstract: | Disk subsystem is known to be a major contributor to overall power consumption of high-end parallel systems. Past research proposed several architectural-level techniques to reduce disk power by taking advantage of idle periods experienced by disks. Although such techniques have been known to be effective in certain cases, they share a common drawback: They operate in a reactive manner, i.e., they control disk power by observing past disk activity (for example, idle and active periods) and estimating future ones. Consequently, they can miss opportunities for saving power and incur significant performance penalties due to inaccuracies in predicting idle and active times. Motivated by this observation, this paper proposes and evaluates a compiler-driven approach to reducing disk power consumption of array-based scientific applications executing on parallel architectures. The proposed approach exposes disk layout information to the compiler, allowing it to derive the disk access pattern, i.e., the order in which parallel disks are accessed. This paper demonstrates two uses of this information. First, we can implement proactive disk power management, i.e., we can select the most appropriate power-saving strategy and disk-preactivation strategy based on the compiler-predicted future idle and active periods of parallel disks. Second, we can restructure the application code to increase the length of idle disk periods, which leads to better exploitation of available power-saving capabilities. We implemented both these approaches within an optimizing compiler and tested their effectiveness using a set of benchmark codes from the Spec 2000 suite and a disk power simulator. Our results show that the compiler-driven disk power management is very promising. The experimental results also reveal that, although proactive disk power management is very effective, code restructuring for disk power achieves additional energy savings across all the benchmarks tested, and these savings are very close to optimal savings that can be obtained through an Integer Linear Programming (ILP)-based scheme. [ABSTRACT FROM AUTHOR] |
| Copyright of IEEE Transactions on Parallel & Distributed Systems is the property of IEEE 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: 26373195 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Compiler-Directed Energy Optimization for Parallel-Disk-Based Systems. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Seung+Woo+Son%22">Seung Woo Son</searchLink><relatesTo>1</relatesTo><i> sson@cse.psu.edu</i><br /><searchLink fieldCode="AR" term="%22Guangyu+Chen%22">Guangyu Chen</searchLink><relatesTo>2</relatesTo><i> ozturk@cse.psu.edu</i><br /><searchLink fieldCode="AR" term="%22Ozturk%2C+Ozcan%22">Ozturk, Ozcan</searchLink><relatesTo>1</relatesTo><i> kandemirj@cse.psu.edu</i><br /><searchLink fieldCode="AR" term="%22Kandemir%2C+Mahmut%22">Kandemir, Mahmut</searchLink><relatesTo>1</relatesTo><i> guchen@microsoft.com</i><br /><searchLink fieldCode="AR" term="%22Choudhary%2C+Alok%22">Choudhary, Alok</searchLink><relatesTo>3</relatesTo><i> choudhar@ece.northwestern.edu</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22IEEE+Transactions+on+Parallel+%26+Distributed+Systems%22">IEEE Transactions on Parallel & Distributed Systems</searchLink>. Sep2007, Vol. 18 Issue 9, p1241-1257. 17p. 9 Diagrams, 4 Charts, 11 Graphs. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Optimizing+compilers%22">Optimizing compilers</searchLink><br /><searchLink fieldCode="DE" term="%22Compilers+%28Computer+programs%29%22">Compilers (Computer programs)</searchLink><br /><searchLink fieldCode="DE" term="%22Algorithms%22">Algorithms</searchLink><br /><searchLink fieldCode="DE" term="%22Parallel+algorithms%22">Parallel algorithms</searchLink><br /><searchLink fieldCode="DE" term="%22Linear+programming%22">Linear programming</searchLink><br /><searchLink fieldCode="DE" term="%22Electronic+file+management%22">Electronic file management</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+programming%22">Computer programming</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+storage+devices%22">Computer storage devices</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+industry%22">Computer industry</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Disk subsystem is known to be a major contributor to overall power consumption of high-end parallel systems. Past research proposed several architectural-level techniques to reduce disk power by taking advantage of idle periods experienced by disks. Although such techniques have been known to be effective in certain cases, they share a common drawback: They operate in a reactive manner, i.e., they control disk power by observing past disk activity (for example, idle and active periods) and estimating future ones. Consequently, they can miss opportunities for saving power and incur significant performance penalties due to inaccuracies in predicting idle and active times. Motivated by this observation, this paper proposes and evaluates a compiler-driven approach to reducing disk power consumption of array-based scientific applications executing on parallel architectures. The proposed approach exposes disk layout information to the compiler, allowing it to derive the disk access pattern, i.e., the order in which parallel disks are accessed. This paper demonstrates two uses of this information. First, we can implement proactive disk power management, i.e., we can select the most appropriate power-saving strategy and disk-preactivation strategy based on the compiler-predicted future idle and active periods of parallel disks. Second, we can restructure the application code to increase the length of idle disk periods, which leads to better exploitation of available power-saving capabilities. We implemented both these approaches within an optimizing compiler and tested their effectiveness using a set of benchmark codes from the Spec 2000 suite and a disk power simulator. Our results show that the compiler-driven disk power management is very promising. The experimental results also reveal that, although proactive disk power management is very effective, code restructuring for disk power achieves additional energy savings across all the benchmarks tested, and these savings are very close to optimal savings that can be obtained through an Integer Linear Programming (ILP)-based scheme. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of IEEE Transactions on Parallel & Distributed Systems is the property of IEEE 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=26373195 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1109/TPDS.2007.1056 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 17 StartPage: 1241 Subjects: – SubjectFull: Optimizing compilers Type: general – SubjectFull: Compilers (Computer programs) Type: general – SubjectFull: Algorithms Type: general – SubjectFull: Parallel algorithms Type: general – SubjectFull: Linear programming Type: general – SubjectFull: Electronic file management Type: general – SubjectFull: Computer programming Type: general – SubjectFull: Computer storage devices Type: general – SubjectFull: Computer industry Type: general Titles: – TitleFull: Compiler-Directed Energy Optimization for Parallel-Disk-Based Systems. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Seung Woo Son – PersonEntity: Name: NameFull: Guangyu Chen – PersonEntity: Name: NameFull: Ozturk, Ozcan – PersonEntity: Name: NameFull: Kandemir, Mahmut – PersonEntity: Name: NameFull: Choudhary, Alok IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 09 Text: Sep2007 Type: published Y: 2007 Identifiers: – Type: issn-print Value: 10459219 Numbering: – Type: volume Value: 18 – Type: issue Value: 9 Titles: – TitleFull: IEEE Transactions on Parallel & Distributed Systems Type: main |
| ResultId | 1 |
