On approximating the ideal random access machine by physical machines

ABSTRACT

The capability of the Random Access Machine (RAM) to execute any instruction in constant time is not realizable, due to fundamental physical constraints on the minimum size of devices and on the maximum speed of signals. This work explores how well the ideal RAM performance can be approximated, for significant classes of computations, by machines whose building blocks have constant size and are connected at a constant distance.

A novel memory structure is proposed, which is pipelined (can accept a new request at each cycle) and hierarchical, exhibiting optimal latency a(x) = O(x^1/d) to address x, in d-dimensional realizations.

In spite of block-transfer or other memory-pipeline capabilities, a number of previous machine models do not achieve a full overlap of memory accesses. These are examples of machines with explicit data movement. It is shown that there are direct-flow computations (without branches and indirect accesses) that require time superlinear in the number of instructions, on all such machines.

To circumvent the explicit-data-movement constraints, the Speculative Prefetcher (SP) and the Speculative Prefetcher and Evaluator (SPE) processors are developed. Both processors can execute any direct-flow program in linear time. The SPE also executes in linear time a class of loop programs that includes many significant algorithms. Even quicksort, a somewhat irregular, recursive algorithm admits a linear-time SPE implementation. A relation between instructions called address dependence is introduced, which limits memory-access overlap and can lead to superlinear time, as illustrated with the classical merging algorithm.

REFERENCES

• 1: Abolhassan, F., Drefenstedt, R., Keller, J., Paul, W. J., and Scheerer, D. 1993. On the physical design of prams. Comput. J. 36, 8, 756--762.
• 2: A. Aggarwal , B. Alpern , A. Chandra , M. Snir, A model for hierarchical memory, Proceedings of the nineteenth annual ACM symposium on Theory of computing, p.305-314, January 1987, New York, New York, United States [doi>10.1145/28395.28428]
• 3: Alok Aggarwal , Ashok K. Chandra , Marc Snir, Hierarchical memory with block transfer, Proceedings of the 28th Annual Symposium on Foundations of Computer Science, p.204-216, October 12-14, 1987 [doi>10.1109/SFCS.1987.31]
• 4: Ken Kennedy , John R. Allen, Optimizing compilers for modern architectures: a dependence-based approach, Morgan Kaufmann Publishers Inc., San Francisco, CA, 2001
• 5: Alpern, B., Carter, L., Feig, E., and Selker, T. 1994. The uniform memory hierarchy model of computation. Algorithmica 12, 2/3, 72--109.
• 6: Robert Alverson , David Callahan , Daniel Cummings , Brian Koblenz , Allan Porterfield , Burton Smith, The Tera computer system, Proceedings of the 4th international conference on Supercomputing, p.1-6, June 11-15, 1990, Amsterdam, The Netherlands
• 7: Jack Perdue, Predicting Performance on SMPs. A Case Study: The SGI Power Challenge, Proceedings of the 14th International Symposium on Parallel and Distributed Processing, p.729, May 01-05, 2000
• 8: John Backus, Can programming be liberated from the von Neumann style?: a functional style and its algebra of programs, Communications of the ACM, v.21 n.8, p.613-641, Aug. 1978 [doi>10.1145/359576.359579]
• 9: Gianfranco Bilardi , Kattamuri Ekanadham , Pratap Pattnaik, Computational power of pipelined memory hierarchies, Proceedings of the thirteenth annual ACM symposium on Parallel algorithms and architectures, p.144-152, July 2001, Crete Island, Greece [doi>10.1145/378580.378615]
• 10: Gianfranco Bilardi , Kattamuri Ekanadham , Pratap Pattnaik, Optimal organizations for pipelined hierarchical memories, Proceedings of the fourteenth annual ACM symposium on Parallel algorithms and architectures, August 10-13, 2002, Winnipeg, Manitoba, Canada [doi>10.1145/564870.564886]
• 11: Gianfranco Bilardi , Kattamuri Ekanadham , Pratap Pattnaik, An Address Dependence Model of Computation for Hierarchical Memories with Pipelined Transfer, Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 8, p.202.2, April 04-08, 2005 [doi>10.1109/IPDPS.2005.85]
• 12: Gianfranco Bilardi , Kattamuri Ekanadham , Pratap Pattnaik, The Speculative Prefetcher and Evaluator Processor for Pipelined Memory Hierarchies, Proceedings of the International Workshop on Innovative Architecture for Future Generation High Performance Processors and Systems, p.29-43, January 23-25, 2006 [doi>10.1109/IWIAS.2006.38]
• 13: Gianfranco Bilardi , Enoch Peserico, A Characterization of Temporal Locality and Its Portability across Memory Hierarchies, Proceedings of the 28th International Colloquium on Automata, Languages and Programming,, p.128-139, July 08-12, 2001
• 14: Gianfranco Bilardi , Andrea Pietracaprina , Paolo D'Alberto, On the Space and Access Complexity of Computation DAGs, Proceedings of the 26th International Workshop on Graph-Theoretic Concepts in Computer Science, p.47-58, June 15-17, 2000
• 15: Gianfranco Bilardi , Franco P. Preparata, Horizons of parallel computation, Journal of Parallel and Distributed Computing, v.27 n.2, p.172-182, June 1995 [doi>10.1006/jpdc.1995.1080]
• 16: Bernard Chazelle , Louis Monier, A model of computation for VLSI with related complexity results, Journal of the ACM (JACM), v.32 n.3, p.573-588, July 1985 [doi>10.1145/3828.3834]
• 17: Cook, S. A., and Reckhow, R. A. 1973. Time-bounded random access machines. J. Comput. Syst. Sci. 7, 4, 354--375.
• 18: John Fotheringham, Dynamic storage allocation in the Atlas computer, including an automatic use of a backing store, Communications of the ACM, v.4 n.10, p.435-436, Oct. 1961 [doi>10.1145/366786.366800]
• 19: Matteo Frigo , Charles E. Leiserson , Harald Prokop , Sridhar Ramachandran, Cache-Oblivious Algorithms, Proceedings of the 40th Annual Symposium on Foundations of Computer Science, p.285, October 17-18, 1999
• 20: John L. Hennessy , David A. Patterson, Computer architecture: a quantitative approach, Morgan Kaufmann Publishers Inc., San Francisco, CA, 02
• 21: Hong Jia-Wei , H. T. Kung, I/O complexity: The red-blue pebble game, Proceedings of the thirteenth annual ACM symposium on Theory of computing, p.326-333, May 11-13, 1981, Milwaukee, Wisconsin, United States [doi>10.1145/800076.802486]
• 22: David L. Kuck, Structure of Computers and Computations, John Wiley & Sons, Inc., New York, NY, 1978
• 23: Charles Eric Leiserson, Area-efficient vlsi computation, Carnegie Mellon University, Pittsburgh, PA, 1981
• 24: Luccio, F., and Pagli, L. 1993. A model of sequential computation with pipelined access to memory. Math. Syst. Theory 26, 4, 343--356.
• 25: Mattson, R. L., Gecsei, J., Slutz, D. R., and Traiger, I. L. 1970. Evaluation techniques for storage hierarchies. IBM Syst. J. 9, 2, 78--117.
• 26: Metcalf, C. 1993. Data prefetching: A cost/performance analysis. Area Exam, MIT, Cambridge.
• 27: Velijko Milutinovic , Mateo Valero, Guest Editors' Introduction-Cache Memory and Related Problems: Enhancing and Exploiting the Locality, IEEE Transactions on Computers, v.48 n.2, p.97-99, February 1999 [doi>10.1109/TC.1999.752650]
• 28: Polychronopoulos, C. 1987. Loop coalescing: A compiler transformation for parallel machines. In Proceedings of the International Conference on Parallel Processing. 235--242.
• 29: Steven A. Przybylski, Cache and memory hierarchy design: a performance-directed approach, Morgan Kaufmann Publishers Inc., San Francisco, CA, 1990
• 30: John E. Savage, Models of Computation: Exploring the Power of Computing, Addison-Wesley Longman Publishing Co., Inc., Boston, MA, 1997
• 31: Smith, B. J. 1981. Architecture and applications of the HEP multiprocessor system. In Real-Time Signal Processing IV. vol. 298. 241--248.
• 32: Valiant, L. G. 1990. General purpose parallel architectures. Elsevier-MIT Press, Amsterdam, The Netherlands.
• 33: Jeffrey Scott Vitter, External Memory Algorithms, Proceedings of the 6th Annual European Symposium on Algorithms, p.1-25, August 24-26, 1998
• 34: von Neumann, J. 1945. First draft of a report on the EDVAC. http://www.virtualtravelog.net/enteries/2003-08-TheFirstDraft.pdf.
• 35: Whaley, R. C., and Dongarra, J. J. 1998. Automatically Tuned Linear Algebra Software. http://www.netlib.org/atlas/index.html.
• 36: Maurice V. Wilkes, The memory gap and the future of high performance memories, ACM SIGARCH Computer Architecture News, v.29 n.1, p.2-7, March 2001 [doi>10.1145/373574.373576]
• 37: Michael Joseph Wolfe , Carter Shanklin , Leda Ortega, High Performance Compilers for Parallel Computing, Addison-Wesley Longman Publishing Co., Inc., Boston, MA, 1995
• 38: Wm. A. Wulf , Sally A. McKee, Hitting the memory wall: implications of the obvious, ACM SIGARCH Computer Architecture News, v.23 n.1, p.20-24, March 1995 [doi>10.1145/216585.216588]
• 39: Kamen Yotov , Tom Roeder , Keshav Pingali , John Gunnels , Fred Gustavson, An experimental comparison of cache-oblivious and cache-conscious programs, Proceedings of the nineteenth annual ACM symposium on Parallel algorithms and architectures, June 09-11, 2007, San Diego, California, USA [doi>10.1145/1248377.1248394]

INDEX TERMS

Primary Classification:
  B. Hardware
  B.3 MEMORY STRUCTURES
      B.3.0 General

Additional Classification:
  C. Computer Systems Organization
  C.0 GENERAL
  C.1 PROCESSOR ARCHITECTURES
      C.1.0 General

  F. Theory of Computation
  F.0 GENERAL
  F.1 COMPUTATION BY ABSTRACT DEVICES
      F.1.0 General
  F.2 ANALYSIS OF ALGORITHMS AND PROBLEM COMPLEXITY
      F.2.0 General

General Terms:
Algorithms, Performance, Theory

Keywords:
Physical constraints on machines, pipelined hierarchical memory, speculative processors