Mathematical Programming Computation

, Volume 7, Issue 4, pp 429–469 | Cite as

PEBBL: an object-oriented framework for scalable parallel branch and bound

  • Jonathan Eckstein
  • William E. Hart
  • Cynthia A. Phillips
Full Length Paper

Abstract

Parallel Enumeration and Branch-and-Bound Library (PEBBL) is a C++ class library implementing the underlying operations needed to support a wide variety of branch-and-bound algorithms on MPI-based message-passing distributed-memory parallel computing environments. PEBBL can be customized to support application-specific operations, while managing the generic aspects of branch and bound, such as maintaining the active subproblem pool across multiple processors, load balancing, and termination detection. PEBBL is designed to provide highly scalable performance on large numbers of processor cores. We describe the basics of PEBBL’s architecture, with emphasis on the features most critical to is high scalability, including its flexible two-level load balancing architecture and its support for a synchronously parallel ramp-up phase. We also present an example application: the maximum monomial agreement problem arising from certain machine learning applications. For sufficiently difficult problem instances, we show essentially linear speedup on over 6000 processor cores, demonstrating a new state of the art in scalability for branch-and-bound implementations. We also show how processor cache effects can lead to reproducibly superlinear speedups.

Keywords

Branch and bound Parallel computation 

Mathematics Subject Classification

90C57 65Y05 

References

  1. 1.
    Asuncion, A., Newman, D.J.: UCI machine learning repository (2007). http://www.ics.uci.edu/~mlearn/MLRepository.html
  2. 2.
    Bader, D.A., Hart, W.E., Phillips, C.A.: Parallel algorithm design for branch and bound. In: Greenberg, H.J. (ed.) Tutorials on Emerging Methodologies and Applications in Operations Research, pp. 5-1–5-44. Kluwer Academic, Dordrecht (2004)Google Scholar
  3. 3.
    Benaïchouche, M., Cung, V.D., Dowaji, S., Cun, B.L., Mautor, T., Roucairol, C.: Building a parallel branch and bound library. In: Solving Combinatorial Optimization Problems in Parallel—Methods and Techniques, Lecture Notes in Computer Science, vol. 1054, pp. 201–231. Springer-Verlag, London (1996)Google Scholar
  4. 4.
    Bendjoudi, A., Melab, N., Talbi, E.G.: Fault-tolerant mechanism for hierarchical branch and bound algorithm. In: Workshops Proceedings, 25th IEEE International Parallel and Distributed Processing Symposium, Workshop on Large-Scale Parallel Processing (LSPP), pp. 1806–1814. Anchorage, Alaska (20011)Google Scholar
  5. 5.
    Blelloch, G.E.: Scans as primitive parallel operations. IEEE Trans. Comput. 38(11), 1526–1538 (1989)CrossRefGoogle Scholar
  6. 6.
    Boros, E., Hammer, P.L., Ibaraki, T., Kogan, A.: Logical analysis of numerical data. Math. Program. 79(1–3), 163–190 (1997)MATHMathSciNetGoogle Scholar
  7. 7.
    Brewer, T.: A highly scalable system utilizing up to 128 PA-RISC processors. In: COMPCON, Technologies for the Information Superhighway, Digest of Papers, pp. 133–140 (1995). http://dblp.uni-trier.de/db/conf/compcon/compcon95.html#Brewer95
  8. 8.
    Cataldo, A.M., Whaley, R.C.: Scaling LAPACK panel operations using parallel cache assignment. In: ACM Principles and Practice of Parallel Programming. ACM, New York (2010)Google Scholar
  9. 9.
    Clausen, J.: Branch and bound algorithms—principles and examples. In: Migdalas, A., Pardalos, P., Storøy, S. (eds.) Parallel Computing in Optimization, Applied Optimization, vol. 7, pp. 239–267. Kluwer Academic, Dordrecht (1997)CrossRefGoogle Scholar
  10. 10.
    Clausen, J., Perregaard, M.: On the best search strategy in parallel branch-and-bound: Best-first search versus lazy depth-first search. Ann. Oper. Res. 90, 1–17 (1999)MATHMathSciNetCrossRefGoogle Scholar
  11. 11.
    Danna, E., Fenelon, M., Gu, Z., Wunderling, R.: Generating multiple solutions for mixed integer programming problems. In: Integer Programming and Combinatorial Optimization, Lecture Notes in Computer Science, vol. 4513, pp. 280–294. Springer, Berlin (2007)Google Scholar
  12. 12.
    Demiriz, A., Bennett, K.P., Shawe-Taylor, J.: Linear programming boosting via column generation. Mach. Learn. 46, 225–254 (2002)MATHCrossRefGoogle Scholar
  13. 13.
    Dobkin, D.P., Gunopulos, D., Maass, W.: Computing the maximum bichromatic discrepancy, with applications to computer graphics and machine learning. J. Comp. Syst. Sci. 52(3), 453–470 (1996)MATHMathSciNetCrossRefGoogle Scholar
  14. 14.
    Eckstein, J.: Control strategies for parallel mixed integer branch and bound. In: Supercomputing ’94: Proceedings of the 1994 ACM/IEEE conference on Supercomputing, pp. 41–48. ACM, New York (1994)Google Scholar
  15. 15.
    Eckstein, J.: Parallel branch-and-bound algorithms for general mixed integer programming on the CM-5. SIAM J. Optim. 4(4), 794–814 (1994)MATHMathSciNetCrossRefGoogle Scholar
  16. 16.
    Eckstein, J.: Distributed versus centralized storage and control for parallel branch and bound: mixed integer programming on the CM-5. Comput. Optim. Appl. 7(2), 199–220 (1997)MATHMathSciNetCrossRefGoogle Scholar
  17. 17.
    Eckstein, J.: How much communication does parallel branch and bound need? INFORMS J. Comput. 9(1), 15–29 (1997)MATHMathSciNetCrossRefGoogle Scholar
  18. 18.
    Eckstein, J., Goldberg, N.: An improved branch-and-bound method for maximum monomial agreement. INFORMS J. Comput. 24(2), 328–341 (2012)MathSciNetCrossRefGoogle Scholar
  19. 19.
    Eckstein, J., Hart, W.E., Phillips, C.: Massively parallel mixed-integer programming: algorithms and applications. In: Heroux, M.A., Raghavan, P., Simon, H. (eds.) Parallel Processing for Scientific Computing, chapter 17, pp. 323–340. Based on the 11th SIAM Conference on Parallel Processing for Scientific Computing. SIAM (2006)Google Scholar
  20. 20.
    Eckstein, J., Hart, W.E., Phillips, C.A.: Resource management in a parallel mixed integer programming package. In: Proceedings of the Intel Supercomputer Users Group (1997). Online proceedingsGoogle Scholar
  21. 21.
    Eckstein, J., Phillips, C.A., Hart, W.E.: PICO: An object-oriented framework for parallel branch and bound. In: Inherently parallel algorithms in feasibility and optimization and their applications (Haifa, 2000), Stud. Comput. Math., vol. 8, pp. 219–265. North-Holland, Amsterdam (2001)Google Scholar
  22. 22.
    Eckstein, J., Phillips, C.A., Hart, W.E.: PEBBL: An object-oriented framework for scalable parallel branch and bound. RUTCOR Research Report RRR 9-2013, Rutgers University (2013)Google Scholar
  23. 23.
    Eckstein, J., Phillips, C.A., Hart, W.E.: PEBBL 1.4.1 user’s guide. RUTCOR Research Report RRR 2-2014, Rutgers University (2014)Google Scholar
  24. 24.
    Elf, M., Gutwenger, C., Jünger, M., Rinaldi, G.: Branch-and-cut algorithms for combinatorial optimization and their implementation in ABACUS. In: Computational Combinatorial Optimization, Optimal or Provably Near-Optimal Solutions, pp. 157–222. Springer-Verlag, London (2001)Google Scholar
  25. 25.
    Goldberg, N., Shan, C.: Boosting optimal logical patterns. In: Proceedings of the Seventh SIAM International Conference on Data Mining, pp. 228–236. SIAM, Minneapolis (2007)Google Scholar
  26. 26.
    Graham, R.L., Woodall, T.S., Squyres, J.M.: Open MPI: A flexible high performance MPI. In: Proceedings, 6th Annual International Conference on Parallel Processing and Applied Mathematics. Poznan, Poland (2005)Google Scholar
  27. 27.
    Gropp, W., Lusk, E., Doss, N., Skjellum, A.: High-performance, portable implementation of the MPI Message Passing Interface Standard. Parallel Comput. 22(6), 789–828 (1996)MATHCrossRefGoogle Scholar
  28. 28.
    Hillis, W.D.: The Connection Machine. MIT Press, Cambridge (1985)Google Scholar
  29. 29.
    Ichnowski, J., Alterovitz, R.: Parallel sampling-based motion planning with superlinear speedup. In: IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 1206–1212 (2012)Google Scholar
  30. 30.
    Jünger, M., Thienel, S.: Introduction to ABACUS–a branch-and-cut system. Oper. Res. Lett. 22(2–3), 83–95 (1998)MATHMathSciNetCrossRefGoogle Scholar
  31. 31.
    Jünger, M., Thienel, S.: The ABACUS system for branch-and-cut-and-price algorithms in integer programming and combinatorial optimization. Software Pract. Expert. 30, 1325–1352 (2000)MATHCrossRefGoogle Scholar
  32. 32.
    Karypis, G., Kumar, V.: Unstructured tree search on SIMD parallel computers: a summary of results. In: Supercomputing ’92: Proceedings of the 1992 ACM/IEEE conference on Supercomputing, pp. 453–462. IEEE Computer Society Press, Los Alamitos (1992)Google Scholar
  33. 33.
    Kearns, M.J., Schapire, R.E., Sellie, L.M.: Toward efficient agnostic learning. Mach. Learn. 17(2–3), 115–141 (1994)MATHGoogle Scholar
  34. 34.
    Kim, M., Lee, H., Lee;, J.: A proportional-share scheduler for multimedia applications. In: IEEE International Conference on Multimedia computing and systems ’97, pp. 484–491. IEEE Computer Society Press, Los Alamitos (1997)Google Scholar
  35. 35.
    Koch, T., Ralphs, T., Shinano, Y.: Could we use a millon cores to solve an integer program? Math. Methods Oper. Res. 76, 67–93 (2012)MATHMathSciNetCrossRefGoogle Scholar
  36. 36.
    Ladányi, L., Ralphs, T.K., Trotter Jr., L.E.: Branch, cut, and price: Sequential and parallel. In: Computational combinatorial optimization (Schloß Dagstuhl, 2000), Lecture Notes in Computer Science, vol. 2241, pp. 223–260. Springer, Berlin (2001)Google Scholar
  37. 37.
    Mahanti, A., Daniel, C.J.: A SIMD approach to parallel heuristic search. Artif. Intel. 60, 243–282 (1993)CrossRefGoogle Scholar
  38. 38.
    Mattern, F.: Algorithms for distributed termination detection. Distrib. Comput. 2, 161–175 (1987)CrossRefGoogle Scholar
  39. 39.
    Menouer, T., LeCun, B., Vander-Swalmen, P.: Partitioning methods to parallelize constraint programming solver using the parallel framework BobPP. In: Proceedings of the First International Conference on Computer Science, Applied Mathematics and Applications (ICCSAMA), pp. 117–127. Springer International Publishing, Switzerland (2013)Google Scholar
  40. 40.
    Nemhauser, G.L., Savelsbergh, M.W.P., Sigismondi, G.C.: MINTO, a mixed INTeger optimizer. Oper. Res. Lett. 15(1), 47–58 (1994)MATHMathSciNetCrossRefGoogle Scholar
  41. 41.
    Ralphs, T.K., Ladányi, L., Saltzman, M.J.: Parallel branch, cut, and price for large-scale discrete optimization. Math. Program. Ser. B 98(1–3), 253–280 (2003)MATHCrossRefGoogle Scholar
  42. 42.
    Ralphs, T.K., Ládanyi, L., Saltzman, M.J.: A library hierarchy for implementing scalable parallel search algorithms. J. Supercomput. 28(2), 215–234 (2004)MATHCrossRefGoogle Scholar
  43. 43.
    Rayward-Smith, V.J., Rush, S.A., McKeown, G.P.: Efficiency considerations in the implementation of parallel branch-and-bound. Ann. Oper. Res. 43(1–4), 123–145 (1993)MATHMathSciNetCrossRefGoogle Scholar
  44. 44.
    Schulz, M., Galarowicz, J., Maghrak, D., Hachfeld, W., Montoya, D., Cranford, S.: Open\(|\)Speedshop: an open source infrastructure for parallel performance analysis. Sci. Program. 16(2–3), 105–121 (2008)Google Scholar
  45. 45.
    Shinano, Y.: ParaSCIP and fiberSCIP libraries to parallelize a customized SCIP solver (2014). http://scip.zib.de/workshop/parascip_libraries.pdf. SCIP Workshop 2014
  46. 46.
    Shinano, Y., Achterberg, T., Berthold, T., Heinz, S., Koch, T.: paraSCIP: a parallel extension of SCIP. In: Competence in High Performance Computing 2010, pp. 135–148. Springer-Verlag, Belin Heidelberg (2012)Google Scholar
  47. 47.
    Shinano, Y., Harada, K., Hirabayashi, R.: Control schemes in a generalized utility for parallel branch-and-bound algorithms. In: IPPS ’97: Proceedings of the 11th International Symposium on Parallel Processing, p. 621. IEEE Computer Society, Washington, DC (1997)Google Scholar
  48. 48.
    Shinano, Y., Higaki, M., Hirabayashi, R.: A generalized utility for parallel branch and bound algorithms. In: SPDP ’95: Proceedings of the 7th IEEE Symposium on Parallel and Distributeed Processing, p. 392. IEEE Computer Society, Washington, DC (1995)Google Scholar
  49. 49.
    Snir, M., Otto, S.W., Walker, D.W., Dongarra, J., Huss-Lederman, S.: MPI: The Complete Reference. MIT Press, Cambridge (1995)Google Scholar
  50. 50.
    Tschöke, S., Polzer, T.: Portable parallel branch-and-bound library: PPBB-Lib user manual version 2.0. Tech. rep., Department of Computer Science, University of Paderborn (1996)Google Scholar
  51. 51.
    Waldspurger, C.A.: Lottery and stride scheduling: flexibile proportional-share resource management. Ph.D. thesis, MIT Department of Electrical Engineering and Computer Science, Cambridge (1996)Google Scholar
  52. 52.
    Wu, M.S., Aluru, S., Kendall, R.A.: Mixed mode matrix multiplication. In: Proceedings of the IEEE International Conference on Cluster Computing, pp. 195–203 (2002)Google Scholar
  53. 53.
    Wu, X. (ed.): Performance Evaluation, Prediction and Visualization of Parallel Systems, vol. 4 of The Kluwer International Series on Asian Studies in Computer and Information Science. Kluwer Academic, Springer (1999)Google Scholar
  54. 54.
    Xu, Y., Ralphs, T.K., Ladányi, L., Saltzman, M.J.: Computational experience with a software framework for parallel integer programming. INFORMS J. Comput. 21, 383–397 (2009). http://coral.ie.lehigh.edu/~ted/files/papers/CHiPPS-Rev.pdf

Copyright information

© Springer-Verlag Berlin Heidelberg and The Mathematical Programming Society 2015

Authors and Affiliations

  • Jonathan Eckstein
    • 1
  • William E. Hart
    • 2
  • Cynthia A. Phillips
    • 3
  1. 1.Department of Management Science and Information Systems and RUTCORRutgers UniversityPiscatawayUSA
  2. 2.Center for Computing ResearchSandia National LaboratoriesAlbuquerqueUSA
  3. 3.Center for Computing ResearchSandia National LaboratoriesAlbuquerqueUSA

Personalised recommendations