This Article 
 Bibliographic References 
 Add to: 
Analytical Performance Modeling of Hierarchical Mass Storage Systems
October 1997 (vol. 46 no. 10)
pp. 1103-1118

Abstract—Mass storage systems are finding greater use in scientific computing research environments for retrieving and archiving the large volumes of data generated and manipulated by scientific computations. This paper presents a queuing network model that can be used to carry out capacity planning studies of hierarchical mass storage systems. Measurements taken on a Unitree mass storage system and a detailed workload characterization provided the workload intensity and resource demand parameters for the various types of read and write requests. The performance model developed here is based on approximations to multiclass Mean Value Analysis of queuing networks. The approximations were validated through the use of discrete event simulation and the complete model was validated and calibrated through measurements. The resulting model was used to analyze three different scenarios: effect of workload intensity increase, use of file compression at the server and client, and use of file abstractions.

[1] R.H. Katz, T.E. Anderson, J.K. Ousterhout, and D.A. Patterson, "Robo-Line Storage: Low Latency, High Capacity Storage Systems over Geographically Distributed Networks," Technical Report UCB/S2K-91-3, Univ. of California, Berkeley, Mar. 1994.
[2] E.L. Miller and R.H. Katz, "An Analysis of File Migration in a Unix Supercomputing Environment," Technical Report UCB/CSD-92-712, Univ. of California, Berkeley, Mar. 1993.
[3] A. Tarshish and E. Salmon, "The Growth of the Unitree Mass Storage System at the NASA Center for Computational Sciences," Proc. Third NASA GSFC Conf. Mass Storage Systems and Technologies, pp. 19-21,College Park, Md., Oct. 1993.
[4] S. Coleman and S. Miller, "Mass Storage System Reference Model: Version 4," Proc. Goddard Conf. Mass Storage Systems and Technologies,College-Park, Md., 1992.
[5] K.K. Ramakrishnan and J.S. Emer, "Performance Analysis of Mass Storage Service Alternatives for Distributed Systems," IEEE Trans. Software Eng., vol. 15, no. 2, pp. 120-133, Feb. 1989.
[6] E. Drakopoulos and M.J. Merges, "Performance Study of Client-Server Storage Systems," Proc. 11th IEEE Symp. Mass Storage Systems, pp. 67-72,Monterey, Calif., 1991.
[7] E. Drakopoulos and M.J. Merges, "Performance Analysis of Client-Server Storage Systems," IEEE Trans. Computers, vol. 41, no. 11, pp. 1,442-1,452, Nov. 1992.
[8] O.I. Pentakalos and Y. Yesha, "Evaluating the Effect of Online Data Compression on a Disk Cache of a Mass Storage System," Proc. Fourth Goddard Conf. Mass Storage Systems and Technologies,College Park, Md., Mar. 1995.
[9] P.M. Chen, E.K. Lee, G.A. Gibson, R.H. Katz, and D.A. Patterson, "RAID: High-Performance Reliable Secondary Storage," ACM Computing Surveys, vol. 36, no. 3, pp. 145-185, Aug. 1994.
[10] D.A. Menascé, V.A.F. Almeida, and L.W. Dowdy, Capacity Planning and Performance Modeling: From Mainframes to Client-Server Systems.Englewood Cliffs, N.J.: Prentice Hall, 1994.
[11] L. Kaufman and P.J. Rousseeuw, Finding Groups in Data.New York: John Wiley&Sons, 1990.
[12] H. Schwetman, "CSIM: A C-based, Process Oriented Simulation Language," Proc. 1991 Winter Simulation Conf., pp. 387-396, 1991.
[13] S. Chen and D. Towsley, "A Performance Evaluation of RAID Architectures," Technical Report UM-CS-1992-067, Univ. of Massachusetts, Amherst, 1992.
[14] E.K. Lee and R.H. Katz, "An Analytic Performance Model of Disk Arrays," Proc. 1993 SIGMETRICS Conf. Measurement and Modeling of Computer Systems, pp. 98-109,Santa Clara, Calif., 1993.
[15] R.H. Johnson, "UNIX Metrics: Is the Data in Open Systems the Same from Platform to Platform?" Proc. 1994 Computer Measurement Group Conf., pp. 796-807,Orlando, Fla., 1994.
[16] J. Bleker, "UNIX Capacity Planning Hints and Tips," Proc. 1994 Computer Measurement Group Conf., pp. 1-7,Orlando, Fla., 1994,.
[17] S. Sarawagi and M. Stonebraker, "Single Query Optimization for Tertiary Memory," Technical Report UCB:S2K-94-45, Univ. of California, Berkeley, 1994.
[18] K.S. Trivedi, Probability and Statistics with Reliability, Queuing, and Computer Science Applications. Prentice Hall, 1982.
[19] J.C. Lowery, "Calibration and Predictive Modeling of Computer Systems," PhD thesis, Vanderbilt Univ., Nashville, Tenn., 1992.
[20] J.C. Lowery and L.W. Dowdy, "New Directions in Model Calibration," Proc. CMG '91 Int'l Conf. Management and Performance Evaluation of Computer Systems, Dec. 1991.
[21] The Computer Environments and Research Requirements Committee, "NASA Earth and Space Sciences Computing Requirements for 1997-2004," internal report, Jan. 1995.
[22] O.I. Pentakalos and Y. Yesha, "Online Data Compression for Mass Storage File Systems," Technical Report TR-CS-95-05, Univ. of Maryland Baltimore County, July 1995.
[23] D. Mack, "compress version 4.0," Usenet comp.sources.misc:volume20 /compress.
[24] J. Ziv and A. Lempel, "A Universal Algorithm for Sequential Data Compression," IEEE Trans. Information Theory, vol. 23, no. 3, pp. 337-343, 1977.
[25] J.C. Tilton, M. Manohar, and J.A. Newcomer, "Earth Science Data Compression Issues and Activities," Remote Sensing Reviews, vol. 9, pp. 271-298, Sept. 1994.

Index Terms:
Mass storage systems, queuing network modeling, mean-value analysis, Unitree central file manager, compression, file abstraction.
Odysseas I. Pentakalos, Daniel A. Menascé, Milton Halem, Yelena Yesha, "Analytical Performance Modeling of Hierarchical Mass Storage Systems," IEEE Transactions on Computers, vol. 46, no. 10, pp. 1103-1118, Oct. 1997, doi:10.1109/12.628395
Usage of this product signifies your acceptance of the Terms of Use.