|
| This Article | ||
| ||
| Share | ||
| Bibliographic References | ||
| Add to: | ||
 Digg  Furl  Spurl  Blink  Simpy  Google  Del.icio.us  Y!MyWeb | ||
| Search | ||
| ||
| ASCII Text | x | ||
| R.P. LaRowe, Jr., C.S. Ellis, M.A. Holliday, "Evaluation of NUMA Memory Management Through Modeling and Measurements," IEEE Transactions on Parallel and Distributed Systems, vol. 3, no. 6, pp. 686-701, November, 1992. | |||
| BibTex | x | ||
| @article{ 10.1109/71.180624, author = {R.P. LaRowe, Jr. and C.S. Ellis and M.A. Holliday}, title = {Evaluation of NUMA Memory Management Through Modeling and Measurements}, journal ={IEEE Transactions on Parallel and Distributed Systems}, volume = {3}, number = {6}, issn = {1045-9219}, year = {1992}, pages = {686-701}, doi = {http://doi.ieeecomputersociety.org/10.1109/71.180624}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, } | |||
| RefWorks Procite/RefMan/Endnote | x | ||
| TY - JOUR JO - IEEE Transactions on Parallel and Distributed Systems TI - Evaluation of NUMA Memory Management Through Modeling and Measurements IS - 6 SN - 1045-9219 SP686 EP701 EPD - 686-701 A1 - R.P. LaRowe, Jr., A1 - C.S. Ellis, A1 - M.A. Holliday, PY - 1992 KW - Index TermsNUMA memory management; nonuniform memory access time; shared-memoryarchitectures; parallel programs; experimental DUnX operating system kernel; BBNGP1000; highly parameterized dynamic page placement policy; analytic model; memorysystem performance; local/remote NUMA architecture; approximate mean-value analysistechniques; experimental data; model predictions; dynamic multiple-copy pageplacement; replication/coherency fault errors; remotely referenced; parallel programming; shared memory systems; storage allocation; storage management VL - 3 JA - IEEE Transactions on Parallel and Distributed Systems ER - | |||
Dynamic page placement policies for NUMA (nonuniform memory access time)shared-memory architectures are explored using two approaches that complement eachother in important ways. The authors measure the performance of parallel programsrunning on the experimental DUnX operating system kernel for the BBN GP1000, whichsupports a highly parameterized dynamic page placement policy. They also develop andapply an analytic model of memory system performance of a local/remote NUMAarchitecture based on approximate mean-value analysis techniques. The model isvalidated against experimental data obtained with DUnX while running a syntheticworkload. The results of this validation show that, in general, model predictions are quitegood. Experiments investigating the effectiveness of dynamic page-placement and, inparticular, dynamic multiple-copy page placement the cost of replication/coherency faulterrors, and the cost of errors in deciding whether a page should move or be remotelyreferenced are described.
[1] S. V. Adve and M.D. Hill. "Weak Ordering--A New definition,"Proc. 17th Ann. Int'l Symp. Computer Architecture, IEEE CS Press, June 1990, pp 2-14.
[2] Adve, S., et al., "Comparison of Hardware and Software Cache Coherence Schemes,"Proc. 18th ISCA, 1991, pp. 298-308.
[3] BBN,Inside the Butterfly GP1000, Cambridge, MA, Oct. 1988.
[4] D. Black, "Scheduling and resource management techniques for multiprocessors," Ph.D. dissertation, Carnegie-Mellon Univ., July 1990.
[5] D. Black, A. Gupta, and W-D. Weber, "Competitive management of distributed shared memory," inSpring COMPCON 89 Dig. Papers, 1989, pp. 184-190.
[6] D. Black and D. Sleator, "Competitive algorithms for replication and migration problems," Tech. Rep. CMU-CS-89-201, Carnegie-Mellon Univ., Nov. 1989.
[7] W. Bolosky, M. Scott, and R. Fitzgerald, "Simple but effective techniques for NUMA memory management," inProc. Twelfth ACM Symp. Oper. Syst. Principles, Dec. 1989, pp. 19-31.
[8] W. Bolosky, M. Scott, R. Fitzgerald, R. Fowler, and A. Cox, "NUMA policies and their relationship to memory architecture," inProc. Architectural Support for Programming Languages and Oper. Syst., Apr. 1991, pp. 212-221.
[9] M.-C. Chiang and G. S. Sohi, "Experience with mean value analysis models for evaluating shared bus, throughput-oriented multiprocessors," inProc. SIGMETRICS Int. Symp. Comput. Perform. Modeling, Measurement and Eval., May 1991, pp. 90-100.
[10] A. L. Cox and R. J. Fowler, "The implementation of a coherent memory abstraction on a NUMA multiprocessor: Experiences with Platinum," inProc. Twelfth ACM Symp. Oper. Syst. Principles, Dec. 1989, pp. 32-43.
[11] M. Dubois and C. Scheurich, "Memory access dependencies in shared-memory multiprocessors,"IEEE Trans. Software Eng., vol. 16, no. 6, pp. 660-673, June 1990.
[12] K. Gharachorloo, A. Gupta, and J. Hennessy, "Performance Evaluation of Memory Consistency Models for Shared-Memory Multiprocessors,"Proc. Fourth Int'l Conf. Architectural Support for Programming Languages and Operating Systems, ACM, New York, 1991, pp. 245- 257.
[13] C.C. Howell and D. E. Mularz, "Exception handling in large Ada systems," inProc. Washington Ada Symp., 1991.
[14] M. A. Holliday, "Page table management in local/remote architectures," inProc. 1988 ACM Int. Conf. Supercomput., July 1988. pp. 1-8.
[15] M.A. Holliday, "Reference history, page size, and migration daemons in local/remote architectures," inProc. 3rd Int. Conf. Architectural Support for Programming Languages and Oper. Syst., Boston, MA, Apr. 1989, pp. 104-112.
[16] R. P. LaRowe Jr., M. A. Holliday, and C. S. Ellis, "An analysis of dynamic page placement on a NUMA multiprocessor," inProc 1992 ACM Sigmetrics and Performance '92 Conf. Measurement and Modeling of Comput. Syst., Newport, RI, May 1992, pp. 23-34.
[17] R. P. LaRowe, Jr., "Page placement for nonuniform memory access time (NUMA) shared memory multiprocessors," Ph.D. dissertation, Duke Univ., Mar. 1991.
[18] R.P. LaRowe, Jr., and C.S. Ellis, "Experimental Comparison of Memory Management Policies for NUMA Multiprocessors,"ACM Trans. Computer Systems, Vol. 9, No. 4, Nov. 1991, pp. 319-363.
[19] R. P. LaRowe, Jr. and C. S. Ellis, "OS experimentation and a user community coexist under the DUnX kernel," inProc. 1991 Int. Conf. Parallel Processing, Aug. 1991, pp. II-158-166.
[20] R. P. LaRowe Jr., C. S. Ellis, and L. S. Kaplan, "The robustness of NUMA memory management," inProc. Thirteenth ACM Symp. Oper. Syst. Principles, Oct. 1991, pp. 137-151.
[21] R. P. LaRowe Jr., J. T. Wilkes, and C. S. Ellis, "Exploiting operating system support for dynamic page placement on a NUMA shared memory multiprocessor," inProc. Symp. Principles and Practice of Parallel Programming, Apr. 1991, pp. 122-132.
[22] S. Leutenegger and M. K. Vernon, "A mean-value performance analysis of a new multiprocessor architecture," inProc. ACM SIGMETRICS Conf. Measurement and Modelling of Comput. Syst., May 1988.
[23] K. Li and P. Hudak, "Memory coherence in Shared Virtual Memory systems," inProc. 5th Ann. ACM Symp. on Princ. of Distr. Computing(Calgary, AB, Can.), Aug. 1986, pp. 229-239.
[24] K. Li and R. Schaefer, "A hypercube shared virtual memory system," inProc. 1989 Int. Conf. Parallel Processing, Aug. 1989, pp. I-125-132.
[25] J. Ramanathan and L. M. Ni, "Critical factors in NUMA memory management," inProc. Eleventh Int. Conf. Distributed Comput. Syst., May 1991, pp. 500-507.
[26] C. Scheurich and M. Dubois, "Dynamic page migration in multiprocessors with distributed global memory," inProc. Eighth Int. Conf. Distributed Comput. Syst., June 1988, pp. 162-169.
[27] J. Torrellas, J. Hennessy, and T. Weil, "Analysis of critical architectural and program parameters in a hierarchical shared-memory multiprocessor," inProc. 1990 ACM Sigmetrics Conf. Measurement and Modeling of Comput. Syst., 1990, pp. 163-172.
[28] M. K. Vernon, E.D. Lazowska, and J. Zahorjan, "An accurate and efficient performance analysis technique for multiprocessor snooping cache-consistency protocols," inProc. 15th Annu. Int. Symp. Comput. Architecture, Honolulu, HI, May 1988, pp. 308-315.
[29] M. K. Vernon, R. Jog, and G. S. Sohi, "Performance analysis of hierarchical cache-consistent multiprocessors,"Perform. Eval., vol. 9, pp. 287-302, 1989.