Article

Free access

The impact of spatial layout of jobs on parallel I/O performance

Authors:

Jens Mache,

Virginia Lo,

Marilynn Livingston,

Sharad GargAuthors Info & Claims

IOPADS '99: Proceedings of the sixth workshop on I/O in parallel and distributed systems

Pages 45 - 56

https://doi.org/10.1145/301816.301830

Published: 01 May 1999 Publication History

PDF eReader

References

[1]

M. Bae and B. Bose. Resource placement in torus-based networks. IEEE 2qeansactions on Computers, 46(10):1083-1092, October 1997.

Digital Library

Google Scholar

[2]

David H. Bailey, Eric Barszcz, Leonardo Dagum, and Horst D. Simon. NAS parallel benchmark results. IEEE Parallel and Distributed Technologll, 1(1):43-51, February 1993.

Digital Library

Google Scholar

[3]

Sandra Johnson Baylor, Caroline Benveniste, and Yarsun Hsu. Performance evaluation of a massively parallel I/O subsystem. In Ravi Join, John Werth, and James C. Browne, editors, Input/Output in Parallel and Distributed Computer S116terns, volume 362 of The Kluwer International Series in Engineering and Computer Science, chapter 13, pages 293-311. Kluwer Academic Publishers, 199t}.

Google Scholar

[4]

S. H. Bokhnri and D. M. Nicol. Balancing contention and synchronization on the Intel Paragon. IEEE Concurrency, 5(2):74- 83, 1997.

Digital Library

Google Scholar

[5]

Y. Cho, M. Winslett, M. Subrarnaniam, Y. Chen, S. Kuo, and K. E. Seamons. Exploiting local data in parallel array I/O on a practical network of workstations. In Proceedings of the 5th Workshop on {/0 in Parallel and Distributed Systems, SC '97, pages 1-13, 1997.

Digital Library

Google Scholar

[6]

S. P. Dandamudi and H. Hadavi. Performance impact of I/O on sender iniatiated and receiver initiated adaptive load sharing in distributed systems. In Proceedings of the International Conference on Parallel and Distributed Computing Syttern#, 1996.

Google Scholar

[7]

Juan Miguel del Rosario, l#jesh Bordawekar, and Alok Choudhary. Improved parallel I/O via a two-phase run-time access strategy. In Proceedings of the IPPS '93 Workshop on Input/Output in Parallel Computer Systems, pages 56-70, Newport Beach, CA, 1993. Also published in Computer Architecture News 21(5), December 1993, pages 31-38.

Digital Library

Google Scholar

[8]

Andreas Eberhart and Jingke Li. Contention-free communication scheduling on 2D meshes. In Proceedings International Conference on Parallel Processing, 1996.

Google Scholar

[9]

Dong Chen et. al. QCDSP machines: Design, performance and cost. http://www.supercomp.org/sc98/awards/#Gordon Bell.

Google Scholar

[10]

D.G. Feitelson. A survey of scheduling in multiprogrammed parallel systems. Technical Report RC 19790 (87657), IBM Research Division, T.J. Watson Research Center, Yorktown Heights, NY 10598, October 1994.

Google Scholar

[11]

Dror G. Feitelson, Peter F. Corbett, Sandra Johnson Ba~1or, and Yarson Hsu. Parallel I/O subsystems in massively parallel supercomputers. IEEE Parallel and Distributed Technology, 3(3):33-47, Fall 1995.

Digital Library

Google Scholar

[12]

Sharad Garg. TFLOPS PFS: Architecture and design of a highly efficient parallel file system. In Proceedings of Supercomputing '98, 1998.

Digital Library

Google Scholar

[13]

Sharad Garg, Robert Godley, Richard Griffiths, Andrew Pfiffer, Terry Prickett, David Robboy, Stan Smith, T. Mack Stallcup, and Stephen Zeisset. Achieving large scale parallelism through operating system resource management on the Intel TFLOPS supercomputer. Intel Technology Journal, Ist quarter 1998.

Google Scholar

[14]

Garth A. Gibson, Jeffrey Scott Vitter, and John Wilkes. Strategic directions in storage I/O issues in large-scale computing. ACM Computing Surveys, 28(4):779-793, December 1996.

Digital Library

Google Scholar

[15]

Ravl Join, John Werth, and James C. Browne, editors. Input/Output in Parallel and Distributed Computer Sllsteras, volume 362 of The Kiuwer International Series in Engineering and Computer Science. Kluwer Academic Publishers, 1996.

Digital Library

Google Scholar

[16]

C. Koelbel, D. Loveman, R. Schreiber, G. Steele, and M. Zosel. High Performance Fortran Handbook. MIT Press, 1994.

Digital Library

Google Scholar

[17]

David Kotz. Disk-directed I/O for MIMD multiprocessors. In Proceedings of the 199# Symposium on Operating Systems Design and Implementation, pages 61-74. USENIX Association, November 1994. Updated as Dartmouth TR PCS-TR94-226 on November 8, 1994.

Digital Library

Google Scholar

[18]

Tomas Lmag and Lance Kurisaki. Nonuniform traffic spots (nuts) in multistage interconnection networks. Journal of Parallel and Distributed Computing, 10:55-67, 1990.

Digital Library

Google Scholar

[19]

Walter Lee, Matthew Frank, Victor Lee, Kenneth Mackenzie, and Larry Rudolph. Implications of I/O for gang scheduled workloads. In Proceedings of the 3nd Workshop on Job Scheduling Strategies .for Parallel Processing, IPPS '97, 1997.

Digital Library

Google Scholar

[20]

K. Li and K.-H. Cheng. A two-dimensional buddy system for dynamic resource allocation in a partitionable mesh connected system. Journal of Parallel and Distributed Computing, 12:79- 83, 1991.

Digital Library

Google Scholar

[21]

Marilynn Livingston and Quentin Stout. Distributing resources in hypercube computers. In Proceedings o/ the Srd Conference on Hypercube Concurrent Cornputer$ and Applications, 1988.

Digital Library

Google Scholar

[22]

V. M. Lo, K. Windisch, W. Liu, and B. Nitzberg. Non-contiguous processor allocation algorithms for mesh-connected multicomputers. IEEE Transactions on Parallel and Distributed Systems, 8(?):712-726, July 1997.

Digital Library

Google Scholar

[23]

Jens Mache. Parallel I/0- and Communication-Sensitive Scheduling on High-Performance Parallel Computers. PhD thesis, Department of Computer and Information Science, University of Oregon, March 1999.

Digital Library

Google Scholar

[24]

Jess Mache and Virginia Lo. The effects of dispersal on messagepassing contention in processor allocation strategies. In Proceedings of the 3rd Joint Conference on Information Sciences, Volume 3, Sessions on Parallel and Distributed Processing, 1997.

Google Scholar

[25]

Jess Maehe, Virginia Lo, and Kurt Windisch. Minimizing message-passing contention in fragmentation-free processor allocation. In Proceedings of the l Oth International Conference on Parallel and Distributed Computing Systems, 1997.

Google Scholar

[26]

Timothy G. Mattson, David Scott, and Stephen Wheat. A TeraFLOP supercomputer in 1996: the ASCI TFLOP system. In Proceedigs of the International Parallel Processing Symposium, 1996.

Digital Library

Google Scholar

[27]

D. Min and M. W. MuSks. A multipath contention model for analyzing job interaction in 2-D mesh multicomputers. In Proceedings of the 8th International Parallel Processing Symposium, pages 744-751# April 1994.

Digital Library

Google Scholar

[28]

Bernd Mohr. Processor allocation on the Cra~ T3E. Personal communication, 1998.

Google Scholar

[29]

S. Q. Moore and L. M. Ni. The effects of network contention on processor allocation strategies. In Proceedings of the l Oth International Parallel Processing Symposium, pages 268-274, 1996.

Digital Library

Google Scholar

[30]

Csaba A. Moritz and Matthew I. Frank. LoGPC: Modeling network contention in message-passing programs. In Proceedings of Sigmetrie$, 1998.

Digital Library

Google Scholar

[31]

L. M. Ni and P. K. McKinley. A survey of wormhole routing techniques in direct networks. IEEE Tr#nJactions on Computers, pages 62-76, February 1993.

Digital Library

Google Scholar

[32]

Bill Nitzberg. Processor allocation on the Intel Paragon. Personal communication, 1998.

Google Scholar

[33]

Bill Nitzberg and Virginia Lo. Collective buffering: Improving parallel I/O performance. In Proceedings of the Si=th IEEE International Symposium on High Performance Distributed Computing, pages 148-157, Portland, OPt, August 1997. IEEE Computer Society Press.

Digital Library

Google Scholar

[34]

William J. Nitzberg. Collective Parallel I/0. PhD thesis, Department of Computer and Information Science, University of Oregon, December 1995.

Digital Library

Google Scholar

[35]

James C. T. Pool and Paul MeMina. Scalable I/O initiative. http://www.cacr.caltech.edu/SIO/.

Google Scholar

[36]

P. Ramananthan and S. Cha}asani. Resource placement with multiple adjacency constraints in k-ary n-cubes. IEEE Ts'ansaction# Par#Uel arid Diatr#buted Systems, 6(5):511-519, May 1995.

Digital Library

Google Scholar

[37]

A.L.N. Reddy, P. Banerjee, and S. G. Abraham. I/o embedding in hypercubes, pages 331-338, August 1988.

Google Scholar

[38]

Emilia Rosti, Giuseppe Serazzi, Evgenia Smirni, and Mark S. Squillante. The impact of I/O on program behavior and parallel scheduling. In Proceedings of Sigme#ricw, 1998.

Digital Library

Google Scholar

[39]

K. E. Seamonl, Y. Chen, P. Jones, J. Jozwiak, and M. Winslett. Server-directed coliective I/O in Panda. In Proceedings o} Super. computing '95, San Diego, CA, December 1995. IEEE Computer Society Press.

Digital Library

Google Scholar

[40]

K. Suzaki, H. Tanuma, S. Hirano, Y. Ichisugi, C. Connelly, and M. Tukamoto. Multi-tasking method on parallel computers which combines a contiguous and a non-contiguous processor partitioning algorithm. In Lecture No#em in Computer Science 118,L pages 641-650, 1996.

Crossref

Google Scholar

[41]

N. F. Tzeng and G. L. Feng. Resource allocation in cube network systems ba#ed on the covering radius. IEEE T#an#actions P#rall~l and Distributed Stlsterns, 7(4):328-342, April 1996.

Digital Library

Google Scholar

[42]

M. Wan, 1t. Moore, G. Kremenek, and K. Steube. A batch scheduler for the Intel Paragon MPP system with a non-contiguous node allocation algorithm. In Proceedings of the #.nd Woricmhop on Job Scheduling Strategies .for Parallel Proce#sin9, IPPB '96, 1996.

Digital Library

Google Scholar

[43]

K. Windisch, V. Lo, D. Feitelson, B. Nitzberg, and 1t. Moore. A comparison of workload traces from two production parallel machines. In Proceedings of the Sizth Symposium on #he Frontiers o/ Massivell/ P#raUel Gomputation# 1996.

Digital Library

Google Scholar

[44]

K. Windisch, J. V. Miller, and V. Lo. ProcSimity: an experimental tool for processor allocation and scheduling in highly parallel systems. In Proceedings of the Fifth Symposium on the Frontiers of Massiveltt Parallel Computation, pages 414-421, 1995.

Digital Library

Google Scholar

[45]

Y. Zhu. Efficient processor allocation strategies for meshconnected parallel computers. Journal of ParaUe| and Distributed Computing, 16:328-337, 1992.

Google Scholar

Cited By

View all

Weinberg JSnavely A(2007)Symbiotic Space-Sharing on SDSC’s DataStar SystemJob Scheduling Strategies for Parallel Processing10.1007/978-3-540-71035-6_10(192-209)Online publication date: 2007
https://doi.org/10.1007/978-3-540-71035-6_10
Weinberg JSnavely A(2006)Symbiotic space-sharing on SDSC's datastar systemProceedings of the 12th international conference on Job scheduling strategies for parallel processing10.5555/1757044.1757054(192-209)Online publication date: 26-Jun-2006
https://dl.acm.org/doi/10.5555/1757044.1757054
Weinberg JSnavely AEgan GMuraoka Y(2006)User-guided symbiotic space-sharing of real workloadsProceedings of the 20th annual international conference on Supercomputing10.1145/1183401.1183450(345-352)Online publication date: 28-Jun-2006
https://dl.acm.org/doi/10.1145/1183401.1183450
Show More Cited By

Index Terms

The impact of spatial layout of jobs on parallel I/O performance

Recommendations

Scheduling parallel jobs to minimize the makespan

We consider the NP-hard problem of scheduling parallel jobs with release dates on identical parallel machines to minimize the makespan. A parallel job requires simultaneously a prespecified, job-dependent number of machines when being processed. We ...
Parallel Algorithms for VLSI Layout Verification

Layout verification determines whether the polygons that represent different mask layers in the chip conform to the technology specifications. Commercial layout verification programs can take tens of hours to run in the flattened representations for ...
Interpreting Parallel Processor Performance Measurements

This paper discusses execution time versus number of simultaneous operations in parallel computing systems. The main focus is on shared memory multiprocessors. A model for execution time as a function of the number of processes used in a computation is ...

Comments

Information & Contributors

Information

Published In

IOPADS '99: Proceedings of the sixth workshop on I/O in parallel and distributed systems

May 1999

98 pages

ISBN:1581131232

DOI:10.1145/301816

Chairmen:
Thomas H. Cormen
Dartmouth College, Hanover, NH
,
Rajeev Thakur
Argonne National Lab, Argonne, IL
,
Daniel A. Reed
Univ. of Illinois at Urbana-Champaign, Urbana

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 01 May 1999

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Qualifiers

Article

Conference

IOPADS99

Sponsor:

IOPADS99: 6th Workshop on I/O in Parallel and Distributed Systems

May 5, 1999

Georgia, Atlanta, USA

Acceptance Rates

IOPADS '99 Paper Acceptance Rate 9 of 30 submissions, 30%;

Overall Acceptance Rate 19 of 56 submissions, 34%

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

5
Total Citations
View Citations
314
Total Downloads

Downloads (Last 12 months)15
Downloads (Last 6 weeks)3

Reflects downloads up to 10 Aug 2024

Other Metrics

View Author Metrics

Citations

Cited By

View all

Weinberg JSnavely A(2007)Symbiotic Space-Sharing on SDSC’s DataStar SystemJob Scheduling Strategies for Parallel Processing10.1007/978-3-540-71035-6_10(192-209)Online publication date: 2007
https://doi.org/10.1007/978-3-540-71035-6_10
Weinberg JSnavely A(2006)Symbiotic space-sharing on SDSC's datastar systemProceedings of the 12th international conference on Job scheduling strategies for parallel processing10.5555/1757044.1757054(192-209)Online publication date: 26-Jun-2006
https://dl.acm.org/doi/10.5555/1757044.1757054
Weinberg JSnavely AEgan GMuraoka Y(2006)User-guided symbiotic space-sharing of real workloadsProceedings of the 20th annual international conference on Supercomputing10.1145/1183401.1183450(345-352)Online publication date: 28-Jun-2006
https://dl.acm.org/doi/10.1145/1183401.1183450
Liu PHsu CWu J(2005)I/O Processor Allocation for Mesh Cluster ComputersProceedings of the 11th International Conference on Parallel and Distributed Systems - Volume 0110.1109/ICPADS.2005.169(105-111)Online publication date: 20-Jul-2005
https://dl.acm.org/doi/10.1109/ICPADS.2005.169
Mache JLo VGarg S(2000)Job scheduling that minimizes network contention due to both communication and I/OProceedings 14th International Parallel and Distributed Processing Symposium. IPDPS 200010.1109/IPDPS.2000.846022(457-463)Online publication date: 2000
https://doi.org/10.1109/IPDPS.2000.846022

View Options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Cited By

Index Terms

Recommendations

Scheduling parallel jobs to minimize the makespan

Parallel Algorithms for VLSI Layout Verification

Interpreting Parallel Processor Performance Measurements