This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Performance Modelling and Comparisons of Global Shared Buffer Management Policies in a Cluster Environment
November 1994 (vol. 43 no. 11)
pp. 1281-1297

Clustering multiple computing nodes has become increasingly popular for reasons of capacity, availability and cost. One approach to clustering is the data sharing approach where a number of loosely coupled nodes share a common database. In this environment, a global shared buffer can be introduced to alleviate the multisystem invalidation effect either as a disk cache or shared intermediate memory. We develop an analytic model to evaluate different shared buffer management policies (SBMPs) which differ in their choice of data granules to be put into the shared buffer. The methodology analyzes all policies using a uniform framework by decomposing the input stream to the shared buffer into multiple (three) component streams based on their effects on the dependency between the private and shared buffer contents. This approach simplifies the problem of analyzing different SBMPs into 1) estimating the rate of each component stream, and 2) evaluating the impact of dependency on each type of component stream and hence the shared buffer hit probability. A detailed simulation model is also developed to validate the analytic model. We also illustrate how the analytic buffer model can be integrated with other system submodels to examine trade-offs between the SBMPs and to estimate optimal shared buffer allocations from a cost-performance point of view.

[1] M. J. Carey, M. J. Franklin, M. Livny, and D. J. Shekita, "Data caching trade-offs in client-server DBMS architectures," inACM SIGMOD, Denver, CO, May 1991, pp. 357-366.
[2] R. I. Casas and K. C. Sevcik, "A buffer management model for use in predicting overall database system performance," inFifth Int. Conf. Data Eng., Los Angeles, CA, Feb. 1989, pp. 463-469.
[3] David Cheriton, "The V Distributed System,"Comm. ACM, Vol. 31, No. 3, Mar. 1988, pp. 314-333.
[4] D.R. Cheriton, H.A. Goosen, and P.D. Boyle, "Multi-Level Shared Caching Techniques for Scalability in VMP-MC,"Proc. 16th Int'l Symp. Computer Architecture, CS Press, Los Alamitos, Calif., May 1989, pp. 16-24.
[5] A. Dan, "Performance analysis of data sharing environments," inACM Distinguished Dissertation Series. Cambridge, MA: MIT Press, 1992.
[6] A. Dan and D. Towsley, "An approximate analysis of the LRU and FIFO buffer replacement schemes," inACM SIGMETRICS, Denver, CO, May 1990, pp. 143-152.
[7] A. Danet al., "The effect of skewed data access on buffer hits and data contention in a data sharing environment," inProc. 16th Int. Conf. Very Large Data Bases, pp. 419-431, Aug. 1990.
[8] A. Dan, D. M. Dias, and P. S. Yu, "Modelling a hierarchical buffer for the data sharing environment," inACM SIGMETRICS, San Diego, CA, May 1991, pp. 156-167.
[9] A. Dan and P. S. Yu, "Performance analysis of buffer coherency policies in a data sharing environment,"IEEE Trans. Parallel Distrib. Syst., vol. 4, no. 3, pp. 289-305, Mar. 1993.
[10] A. Dan, P. S. Yu, and J.-Y. Chung, "Database access characterization for buffer hit prediction," inProc. 9th Int. Conf. Data Eng., Vienna, Austria, Apr., 1993, pp. 134-143.
[11] A. Dan, D. M. Dias, and P. S. Yu, "Buffer analysis for a data shaving environment with skewed data access,"IEEE Trans. Knowledge Data Eng., vol. 6, no. 2, pp. 331-337, Apr. 1994.
[12] D.M. Dias, B. R. Iyer, and P. S. Yu, "Trade-offs between coupling small and large processors for transaction processing,"IEEE Trans. Comput., vol. C-37, no. 3, pp. 310-320, Mar. 1988.
[13] D. M. Dias, B. R. Iyer, J. T. Robinson, and P. S. Yu, "Integrated concurrency-coherency controls for multisystem data sharing,"IEEE Trans. Software Eng., vol. 15, no. 4, pp. 437-448, Apr. 1989.
[14] M. J. Franklin, M. J. Carey, and M. Livny, "Global memory management in client server DBMS architectures," in18th Int. Conf. Very Large Databases, Vancouver, Canada, Aug. 1992, pp. 596-609.
[15] J. P. Gelb, "System managed storage,"IBM Syst. J., vol. 28, no. 1, 1989.
[16] N. P. Kronenberg, H. M. Levy, and W. D. Strecker, "VAXcluster: A closely-coupled distributed system,"ACM Trans. Comput. Syst., vol. 4., no. 2, pp. 130-146, May 1986.
[17] S.S. Lavenberg,Computer Performance Modeling Handbook, Academic Press, New York, 1983.
[18] B. McNutt, "DASD configuration planning: three simple checks," inCMG Conf. Proc., 1988.
[19] C. Mohan and I. Narang, "Recovery and coherency control protocols for fast intersystem page transfer and fine granularity locking in a shared disks transaction environment," inProc. 17th Int. Conf. Very. Large Data Bases, Barcelona, Spain, Sept. 1991, pp. 193-207.
[20] C. Mohan and I. Narang, "Efficient locking and caching of data in the multi-system shared disks transaction environment,"Proc. Int. Conf. Extending Database Technol., Vienna, Mar. 1992.
[21] D.J. Makaroff and D. L. Eager, "Disk cache performance for distributed systems,"10th Int. Conf. Distrib. Computing Syst., Paris, France, May 1990, pp. 212-219.
[22] E. Rahm, "Design of optimistic methods for concurrency control in database sharing systems," inSeventh Int. Conf. Distrib. Computing Syst., Berlin, Germany, September 1987, pp. 154-161.
[23] E. Rahm, "Use of global extended memory for distributed transaction processing," in4th Int. Workshop on High Perform. Trans. Processing, Asilomar, CA, Sept. 1991.
[24] J. P. Strickland, P. P. Uhrowczik, and V. L. Watts, "IMS/VS: An evolving system,"IBM Syst. J., vol. 21, pp. 490-510, 1982.
[25] P. S. Yu, D. M. Dias, J. T. Robinson, B. R. Iyer, and D. W. Cornell, "On coupling multi-systems through data sharing," inProc. IEEE, vol. 75, no. 5, pp. 573-587, May 1987.
[26] K. Barker and M.T. Ozzu, "Reliable Transaction Execution in Multidatabase Systems,"Proc. First Int'l Workshop Interoperability Multidatabase Systems, IEEE CS Press, Los Alamitos, Calif., Order No. 2205, 1991, pp. 344-347.
[27] P. S. Yu and D. M. Dias, "Analysis of hybrid concurrency control schemes for a high data contention environment,"IEEE Trans. Software Eng., vol. 18, no. 2, pp. 118-129, Feb. 1992.
[28] P. S. Yu and A. Dan, "Effect of system dynamics on coupling architectures for transaction processing,"IEEE Parallel Distrib. Syst., vol. 5, no. 12, pp. 139-153, Feb. 1994.
[29] P. S. Yu, and A. Dan, "Impact of workload partitionability on the performance of coupling architectures for transaction processing," inProc. 4th IEEE Symp. Parallel Distrib. Processing, Dallas, TX, Dec. 1992, pp. 40-49.
[30] P. S. Yu and A. Dan, "Performance analysis of affinity clustering on transaction processing coupling architectures," to appear inIEEE Trans. Knowledge Data Eng.
[31] IBM Corp., "Sysplex overview: Introducing data sharing and parallelism in a sysplex," GC28-1208-00, Apr. 1994.

Index Terms:
buffer storage; storage management; transaction processing; performance evaluation; shared memory systems; global shared buffer management policies; cluster environment; global shared buffer; multisystem invalidation effect; disk cache; shared intermediate memory; shared buffer management policies; data granules; component stream; shared buffer hit probability; simulation model; optimal shared buffer allocations; cost-performance; transaction processing; high bandwidth interconnection network; OLTP.
Citation:
A. Dan, P.S. Yu, D.M. Dias, "Performance Modelling and Comparisons of Global Shared Buffer Management Policies in a Cluster Environment," IEEE Transactions on Computers, vol. 43, no. 11, pp. 1281-1297, Nov. 1994, doi:10.1109/12.324561
Usage of this product signifies your acceptance of the Terms of Use.