ACM Home Page
Please provide us with feedback. Feedback
Digital Library logoTake a look at the new version of this page: [ beta version ]. Tell us what you think.
Hierarchical Dynamics, Interarrival Times, and Performance
Full text PdfPdf (548 KB)
Source Conference on High Performance Networking and Computing archive
Proceedings of the 2003 ACM/IEEE conference on Supercomputing table of contents
Page: 28  
Year of Publication: 2003
ISBN:1-58113-695-1
Authors
Stephen D. Kleban  Sandia National Laboratories, Albuquerque, NM
Scott H. Clearwater  Netcom
Sponsor
SIGARCH: ACM Special Interest Group on Computer Architecture
Publisher
IEEE Computer Society  Washington, DC, USA
Bibliometrics
Downloads (6 Weeks): 1,   Downloads (12 Months): 7,   Citation Count: 1
Additional Information:

abstract   references   cited by   index terms   collaborative colleagues  

Tools and Actions: Review this Article  

ABSTRACT

We report on a model of the distribution of job submission interarrival times in supercomputers.Interarrival times are modeled as a consequence of a complicated set of decisions between users, the queuing algorithm, and other policies.This cascading hierarchy of decision-making processes leads to a particular kind of heavy-tailed distribution.Specifically, hierarchically constrained systems suggest that fatter tails are due to more levels coming into play in the overall decision-making process.The key contribution of this paper is that heavier tials resulting from more complex decision-making processes, that ismore hierarchical levels, will lead to overall worse performance, even when the average interarrival time is the same.Finally, we offer some suggestions for how to overcome these issues and the tradeoffs involved.


REFERENCES

Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.

 
1
[1] http://www.llnl.gov/asci/
 
2
[2] J. Beran, R. Sherman, M. S. Taqqu, and W. Willinger, "Long-Range Dependence in variable-Bit-Rate Video Traffic", IEEE Transactions on communication, vol. 43, no. 2/3/4 Feb./Mar./Apr. 1995, pp. 1566-1579.
 
3
 
4
[4] DPCS Manual (unpublished) and Moe Jette and Bob Wood private communication, 2001-2002.
 
5
[5] D. Feitelson, "Sensitivity of parallel Job Scheduling to Fat-Tailed Distribution", Technical Report 2000-44, Inst. Computer Science, The Hebrew University of Jerusalem, Oct 2000 and http://www/cs/huji/ac.il/-feit/pub.html.
 
6
[6] U. Frisch and D. Sornette, "Extreme Deviations and Application", Phys. I France, vol. 7, 1997, pp. 1155-1171.
7
 
8
[8] M. Harchol-Balter, "The Effect of Heavy-Tailed Job Size Distribution on Computer System Design", Proceedings of ASA-IMS Conference on Application of Heavy Tailed Distribution in Economics, Engineering and Statistics, June, 1999.
9
 
10
[10] J. Klafter and M. F. Shelsigner, "On the relationship among three theories of relaxation in disordered systems", Proceeding of the National Academy of Science, vol. 83, Feb. 1986, pp. 848-851.
 
11
 
12
[12] S. Kleban and S. Clearwater, "A Market-Based Architecture for Supercomputer Resource Management", Proceedings of the ACM/IEEE Super Computing 2001 (SC-2001). 2001.
 
13
[13] R. Kohlrausch, "Theories des electrischen Rückstandes in der Leidner Flasche", Pogg. Ann Phys., vol. 91, 1854, pp. 179- 214 (in German).
 
14
[14] J. Laherrere and D. Sornette, "Stretched exponential distributions in Nature and Economy: 'Fat tails' with characteristic tails", European Physics Journal B, vol. 2, 1998, pp. 525-539.
 
15
[15] "LSF Batch Administrator's Guide, Sixth Edition", Platform Computing Corporation, 1998.
 
16
[16] http://umbc7.umbc.edu/nqs/nqsguide.html.
 
17
[17] R. G. Palmer, D. L. Stein, E. Abrahams, and P. W. Anderson, "Models of Hierarchically Constrained Dynamics for Glassy Systems", Phy. Rev. Lett. vol. 53, no. 10, Sep. 1984, pp. 958-961.
 
18
 
19
 
20
[20] J. Voldman, B.B. Mandelbrot, L.W. Hoevel, J. Knight, and P. Rosenfeld, "Fractal Nature of Software-Cache Interaction", IBM Journal of Research and Development, vol. 27, no. 6, Nov. 1981, pp. 164-170.
 
21
[21] W. Willinger, R. Govindan, S. Jamin, V. Paxson, and S. Shenker, "Scaling phenomena in the Internet: Critically examination criticality", Proceeding of the National Academy of Science, vol. 99, Feb. 2002, pp. 2573-2580.
 
22
[22] W. Willinger, M. Taqqu, and A. Erramilli, "A bibliographical guide to self-similiar traffic and performance modeling for modern high-speed networks", Stochastic Networks, F. P. Kelly, S. Zachary and I. Ziedins (eds.), Oxford University Press, Oxford, 1996, pp. 339-366.


Collaborative Colleagues:
Stephen D. Kleban: colleagues
Scott H. Clearwater: colleagues