基于SAGE的分布式高性能优化设计平台
|
摘要
一个通用且高效的优化设计计算平台能有效地缩短飞行器设计周期、降低成本。本文描述了一个基于多学科设计优化(Multidisciplinary design optimization)理念的分布式并行计算框架的设计与实现。结合飞行器设计的实际需求和并行计算的基本原理,我们提出了一种基于分布式流水线的解决方案。该计算框架的特征可以从两个角度阐述:(1)从飞行器设计人员的角度来看,计算框架提供了一个灵活易用的计算模型,使其能将注意力集中于计算逻辑的描述;(2)从计算机科学角度来看,计算框架将复杂的并行计算操作隐藏起来,充分发掘粗粒度并行度,提高计算效率。总的说来,就是提供连接并行计算与飞行器设计专业的桥梁,将计算机科学发展成果以一种易于接受的方式应用于工程设计中。本文讨论管道流水线和任务流水线两种并行计算模型的设计思路,并介绍支持这个计算模型的分布式计算框架的设计思路和实现原理。该计算框架已应用于某飞行器优化设计平台中,并且在中小规模的分布式环境中具有良好的性能和扩展性。
A high performance generic framework can greatly lower the cost of aircraft design. We present the design and implementation of a computing framework based on Multidisciplinary Design Optimization (MDO) theory. According to the requirement and basic principles of parallel computing, we have proposed a distributed pipeline solution. The framework has two major characteristics: 1) From the perspective of aircraft design, the framework made available to aircraft designers a simple and flexible computing model, so that they can concentrate on the description of work flow; 2) From the perspective of computer science, the framework hides the complex of parallel computing and explores the parallelism in computing. In general, we provide a bridge between parallel computing and engineering design. This article introduces two computing models. Design and implementation of underlay computing framework is also discussed. The framework has been applied to one aircraft design platform. Experiments show the good performance and scalability of the framework in medium-scaled and small-scaled applications. .
A high performance generic framework can greatly lower the cost of aircraft design. We present the design and implementation of a computing framework based on Multidisciplinary Design Optimization (MDO) theory. According to the requirement and basic principles of parallel computing, we have proposed a distributed pipeline solution. The framework has two major characteristics: 1) From the perspective of aircraft design, the framework made available to aircraft designers a simple and flexible computing model, so that they can concentrate on the description of work flow; 2) From the perspective of computer science, the framework hides the complex of parallel computing and explores the parallelism in computing. In general, we provide a bridge between parallel computing and engineering design. This article introduces two computing models. Design and implementation of underlay computing framework is also discussed. The framework has been applied to one aircraft design platform. Experiments show the good performance and scalability of the framework in medium-scaled and small-scaled applications. .
引文
[1] AIAA White Paper, Current State of the Art: Multidisciplinary Design Optimization[R]. Washington:AIAA Technical Committee for MDO, 1991.
[2] Sobieszczanski-Sobieski J, Haftka T. Multidisciplinary aerospace design optimization: Survey of Recent Developments[R]. AIAA 96-0711, Nevada: AIAA, 1996.
[3] Balling R J, Sobieszczansk i-Sobieski J. Optimization of coupled systems: A Critical Overview of Approaches[J]. AIAA Journal, 1996, 34(1): 6～17.
[4] Salas A O , Townesnd J C . Framework requirement for MDO application development[A].AIAA/NASA/ISSMO Symposium on Multidisciplinary Analysis and Optimization[C], Sept. 24, 1998, Missouri
[5] Dovi A R, Su P, Murthy T. A distributed object approach to multidisciplinary analysis of aerospace vehicle[A] . AIAA-2000-4898, 8th AIAA/NASA/USAF/ISSMO Symposium on Multidisciplinary Analysis and Optimization Multidisciplinary Analysis and Optimization, 2000
[6] Arora, J.S. 1989: Introduction to optimum design.NewYork: McGraw-Hill
[7] Vanderplaats Research & Development Inc. VisualDOC 3.0: Getting Started Manual, 2002
[8] Eldred M S, DAKOTA G A A. A Multilevel Parallel Object-Oriented Framework for Design Optimization, Parameter Estimation, Uncertainty Quantification, and Sensitivity Analysis[R]. Version 3.0 Users Manual. Sandia Technical Report SAND2001-3796, 2002
[9] Batill S B, Stelmack M A, Yu X Q. Multidisciplinary design optimization of an electric-powered unmanned air vehicle[J]. AircraftDesign, 1999, 2(1): 1—18.
[10] KrooI. Manning V. Collaborative optimization: status and directions[A]. AIAA-2000-4721. 8th AIAA/NASA/USAF/ISSMO Symposium On Multidisciplinary Analysis and Optimization. 2000
[11] AIAA/ASME/ ASCE/AHS/ASC, Structures, Structural Dynamics and Materials Conf., pp. 2724–2732 (held in Kissimmee, FL), Paper No. AIAA-97-1403
[12] Barry Wilkinson, Michael Allen著,陆鑫达译,《并行程序设计》第2版,机械工业出版社,2005年5月
[13] Parallel computing, http://en.wikipedia.org/wiki/Parallel_computing
[14] M. J. Flynn. Some computer organizations and their effectiveness. IEEE Transactions on Computers, C-21(9):948–960, 1972.
[15] D. A. Patterson and J. L. Hennessy. Computer Architecture: A Quantitative Approach. Morgan Kaufmann, San Mateo, CA, 1990.
[16] K. E. Batcher. Design of a massively parallel processor. IEEE Transactions on Computers, 836–840, September 1980.
[17] Blaise Barney, Lawrence Livermore National Laboratory. Introduction to Parallel Computing. https://computing.llnl.gov/tutorials/parallel_comp
[18] Ananth Grama, Anshul Gupta, George Karypis, Vipin Kumar.Introduction to Parallel Computing.Addison Wesley,2003.
[19] David R. Butenhof: Programming with POSIX Threads, Addison-Wesley, ISBN 0-201-63392-2
[20] R. Chandra, R. Menon, L. Dagum, D. Kohr, D. Maydan, J. McDonald, Parallel Programming in OpenMP. Morgan Kaufmann, 2000. ISBN 1558606718
[21] Aoyama, Yukiya; Nakano, Jun (1999) RS/6000 SP: Practical MPI Programming, ITSO
[22] Andrew S. Tanenbaum等.分布式系统原理与范型.清华大学出版社, 2004年9月.
[23]张云勇等.中间件技术原理与应用.清华大学出版社, 2004.
[24] Sun Microsystems, Inc., Java Remote Method Invocation, http://java.sun.com/j2se/1.4.2/docs/guide/rmi/.
[25] Henning, M., and S. Vinoski. Advanced CORBA Programming with C++. Addison-Wesley, 1999.
[26] OMG. The Common Object Request Broker: Architecture and Specification (Revision 3.0.2). Document–formal/2002-12-06, Dec. 2002.
[27] Grimes, R.. Professional DCOM Programming. Chicago, IL: Wrox Press. 1998.
[28] Michi Henning, Mark Spruiell. Distributed Programming with Ice. http://www.zeroc.com, Revision 3.3.0, May 2008.
[29] Xu Li, H.Huang, Wei Shu, Minglu Li and Min-You Wu. VStore: Towards Cooperative Storage in Vehicular Sensor Networks for Mobile Surveillance, IEEE Wireless Communication and Networking Conference, Budapest, Hungary, April2009.
[30] Xu Li, Wei Shu, Minglu Li, Hongyu Huang and Min-You Wu. DTN Routing in Vehicular Sensor Networks, the 51th IEEE Global Communications Conference, New Orleans, LA, USA.
[31]李旭.车载传感器网络的应用及关键技术研究.上海交通大学,2009.
[32] Xu Li, Wei Shu, Minglu Li, et al. Performance Evaluation of Vehicle-based Mobile Sensor Networks for Traffic Monitoring. IEEE Transactions on Vehicular Technology, Volume 58, Issue 4, May 2009 Page(s):1647– 1653.
[33] Xu Li, Wei Shu, Minglu Li, Pei-En Luo, Hongyu Huang and Min-You Wu. Traffic Data Processing in Vehicular Sensor Networks, the 17th International Conference on Computer Communications and Networks, Virgin Islands, USA, August, 2008.
[34]任伶,林新华,伍民友等.上海网格自适应网格引擎.微电子学与计算机, 2007.9期.
[35]陈晓铮,林新华,李明禄,伍民友.“基于SAGE的分布式虚拟现实框架”.《微电子学与计算机》2008年第9期.
[36] M.P. Papazoglou and D. Georgakopoulos. Service-oriented Computing. Communications of the ACM. Volume 46, 2003.
[37] Microsoft, Window Cluster Server Introduction, http://www.windowshpc.net.
[38] Peter S. Pacheco. Parallel Programming with MPI. Morgan Kaufmann, 1997.
[39] D.A.Padua and M.J.Wolfe. Advanced compiler optimizations for supercomputers. In communications of the ACM, volume Volume 29, No.12C, December 1986.
[40] High Performance FORTRAN forum. High Performance FORTRAN specification. Technical report, Rice University, Houston, TX.
[41] W.Richard Stevens, Stephen A.Rago. Advanced Programming in the UNIX Enviroment.
[42] The GStreamer Project. http://gstreamer.freedesktop.org/
[43] I. Foster, C. Kesselman. The Grid: Blueprint for a New Computing Infrastructure. Morgan Kaufman, 1999.
[44] Chatterjee, S., Strosnider, J.K.: Distributed Pipeline Scheduling: A framework for distributed, heterogeneous real-time system design. Comput. J. 38 (1995) 271–285
[45] H. Gonz′alez-V′elez and M. Cole. Towards fully adaptive pipeline parallelism for heterogeneous distributed environments. In ISPA 2006, number 4330 in Lect. Notes Comput.Sc., pages 916–926, Sorrento, Dec. 2006. Springer-Verlag.
[46] Gonzalez-Velez, H.; Cole, M., "An adaptive parallel pipeline pattern for grids," Parallel and Distributed Processing, 2008. IPDPS 2008. IEEE International Symposium on , vol., no., pp.1-11, 14-18 April 2008
[47] L.V.Kale and S.Krishnan. Charm++: A portable concurrent object oriented system based on C++. In Proceedings of OOPSLA-93.,March 1993.(Also:Technical Report UIUCDCS-R-93-1796,March 1993,University of Illinois, Urbana, IL.
[48] Kale L.V. and Ramkumar B. and Sinha A.B and Gursoy A. The CHARM Parallel Programming Language and System: Part I - Description of Language Features. IEEE Transactions on Parallel and Distributed Systems, 1994.
[49] The Cilk Project. http://supertech.csail.mit.edu/cilk/
[50] Cilk: An Efficient Multithreaded Runtime System by Robert D. Blumofe, Christopher F. Joerg, Bradley C. Kuszmaul, Charles E. Leiserson, Keith H. Randall, and Yuli Zhou. Proceedings of the Fifth ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP), pp. 207–216, 1995.
[51] Threading Building Blocks Open Source Project.http://www.threadingbuildingblocks.org/
[52] Reinders, James (2007, July). Intel Threading Building Blocks: Outfitting C++ for Multi-core Processor Parallelism (Paperback) Sebastopol: O'Reilly Media, ISBN 978-0-596-51480-8.
[53] Voss, M. (2006, October). "Demystify Scalable Parallelism with Intel Threading Building Blocks' Generic Parallel Algorithms."
[54]李晓明,闫宏飞,王继民.搜索引擎-原理、技术与系统.科学出版社.2004
[55]叶庆华肖利民祝明发,机群作业管理系统的评价体系,南京大学学报(哲学·人文科学·社会科学),2001年3月
[56] Allan Vermeulen, Gabe Beged-Dov, Patrick Tompsom.The Pipeline Design Pattern. Rogue Wave Software,Inc.1995.
[57] Richard H.Carver, Kuo-Chung Tai. Modern Multithreading. John Wiley & Sons,Inc,2006.
[58] Robert D.Blumofe and Charles E.Leiserson. Scheduling multithreaded computations by work stealing. In Proceedings of the 35th Annual Symposium on Foundations of Computer Science. pages 356-268, Santa Fe, New Mexico, November 1994.
[2] Sobieszczanski-Sobieski J, Haftka T. Multidisciplinary aerospace design optimization: Survey of Recent Developments[R]. AIAA 96-0711, Nevada: AIAA, 1996.
[3] Balling R J, Sobieszczansk i-Sobieski J. Optimization of coupled systems: A Critical Overview of Approaches[J]. AIAA Journal, 1996, 34(1): 6～17.
[4] Salas A O , Townesnd J C . Framework requirement for MDO application development[A].AIAA/NASA/ISSMO Symposium on Multidisciplinary Analysis and Optimization[C], Sept. 24, 1998, Missouri
[5] Dovi A R, Su P, Murthy T. A distributed object approach to multidisciplinary analysis of aerospace vehicle[A] . AIAA-2000-4898, 8th AIAA/NASA/USAF/ISSMO Symposium on Multidisciplinary Analysis and Optimization Multidisciplinary Analysis and Optimization, 2000
[6] Arora, J.S. 1989: Introduction to optimum design.NewYork: McGraw-Hill
[7] Vanderplaats Research & Development Inc. VisualDOC 3.0: Getting Started Manual, 2002
[8] Eldred M S, DAKOTA G A A. A Multilevel Parallel Object-Oriented Framework for Design Optimization, Parameter Estimation, Uncertainty Quantification, and Sensitivity Analysis[R]. Version 3.0 Users Manual. Sandia Technical Report SAND2001-3796, 2002
[9] Batill S B, Stelmack M A, Yu X Q. Multidisciplinary design optimization of an electric-powered unmanned air vehicle[J]. AircraftDesign, 1999, 2(1): 1—18.
[10] KrooI. Manning V. Collaborative optimization: status and directions[A]. AIAA-2000-4721. 8th AIAA/NASA/USAF/ISSMO Symposium On Multidisciplinary Analysis and Optimization. 2000
[11] AIAA/ASME/ ASCE/AHS/ASC, Structures, Structural Dynamics and Materials Conf., pp. 2724–2732 (held in Kissimmee, FL), Paper No. AIAA-97-1403
[12] Barry Wilkinson, Michael Allen著,陆鑫达译,《并行程序设计》第2版,机械工业出版社,2005年5月
[13] Parallel computing, http://en.wikipedia.org/wiki/Parallel_computing
[14] M. J. Flynn. Some computer organizations and their effectiveness. IEEE Transactions on Computers, C-21(9):948–960, 1972.
[15] D. A. Patterson and J. L. Hennessy. Computer Architecture: A Quantitative Approach. Morgan Kaufmann, San Mateo, CA, 1990.
[16] K. E. Batcher. Design of a massively parallel processor. IEEE Transactions on Computers, 836–840, September 1980.
[17] Blaise Barney, Lawrence Livermore National Laboratory. Introduction to Parallel Computing. https://computing.llnl.gov/tutorials/parallel_comp
[18] Ananth Grama, Anshul Gupta, George Karypis, Vipin Kumar.Introduction to Parallel Computing.Addison Wesley,2003.
[19] David R. Butenhof: Programming with POSIX Threads, Addison-Wesley, ISBN 0-201-63392-2
[20] R. Chandra, R. Menon, L. Dagum, D. Kohr, D. Maydan, J. McDonald, Parallel Programming in OpenMP. Morgan Kaufmann, 2000. ISBN 1558606718
[21] Aoyama, Yukiya; Nakano, Jun (1999) RS/6000 SP: Practical MPI Programming, ITSO
[22] Andrew S. Tanenbaum等.分布式系统原理与范型.清华大学出版社, 2004年9月.
[23]张云勇等.中间件技术原理与应用.清华大学出版社, 2004.
[24] Sun Microsystems, Inc., Java Remote Method Invocation, http://java.sun.com/j2se/1.4.2/docs/guide/rmi/.
[25] Henning, M., and S. Vinoski. Advanced CORBA Programming with C++. Addison-Wesley, 1999.
[26] OMG. The Common Object Request Broker: Architecture and Specification (Revision 3.0.2). Document–formal/2002-12-06, Dec. 2002.
[27] Grimes, R.. Professional DCOM Programming. Chicago, IL: Wrox Press. 1998.
[28] Michi Henning, Mark Spruiell. Distributed Programming with Ice. http://www.zeroc.com, Revision 3.3.0, May 2008.
[29] Xu Li, H.Huang, Wei Shu, Minglu Li and Min-You Wu. VStore: Towards Cooperative Storage in Vehicular Sensor Networks for Mobile Surveillance, IEEE Wireless Communication and Networking Conference, Budapest, Hungary, April2009.
[30] Xu Li, Wei Shu, Minglu Li, Hongyu Huang and Min-You Wu. DTN Routing in Vehicular Sensor Networks, the 51th IEEE Global Communications Conference, New Orleans, LA, USA.
[31]李旭.车载传感器网络的应用及关键技术研究.上海交通大学,2009.
[32] Xu Li, Wei Shu, Minglu Li, et al. Performance Evaluation of Vehicle-based Mobile Sensor Networks for Traffic Monitoring. IEEE Transactions on Vehicular Technology, Volume 58, Issue 4, May 2009 Page(s):1647– 1653.
[33] Xu Li, Wei Shu, Minglu Li, Pei-En Luo, Hongyu Huang and Min-You Wu. Traffic Data Processing in Vehicular Sensor Networks, the 17th International Conference on Computer Communications and Networks, Virgin Islands, USA, August, 2008.
[34]任伶,林新华,伍民友等.上海网格自适应网格引擎.微电子学与计算机, 2007.9期.
[35]陈晓铮,林新华,李明禄,伍民友.“基于SAGE的分布式虚拟现实框架”.《微电子学与计算机》2008年第9期.
[36] M.P. Papazoglou and D. Georgakopoulos. Service-oriented Computing. Communications of the ACM. Volume 46, 2003.
[37] Microsoft, Window Cluster Server Introduction, http://www.windowshpc.net.
[38] Peter S. Pacheco. Parallel Programming with MPI. Morgan Kaufmann, 1997.
[39] D.A.Padua and M.J.Wolfe. Advanced compiler optimizations for supercomputers. In communications of the ACM, volume Volume 29, No.12C, December 1986.
[40] High Performance FORTRAN forum. High Performance FORTRAN specification. Technical report, Rice University, Houston, TX.
[41] W.Richard Stevens, Stephen A.Rago. Advanced Programming in the UNIX Enviroment.
[42] The GStreamer Project. http://gstreamer.freedesktop.org/
[43] I. Foster, C. Kesselman. The Grid: Blueprint for a New Computing Infrastructure. Morgan Kaufman, 1999.
[44] Chatterjee, S., Strosnider, J.K.: Distributed Pipeline Scheduling: A framework for distributed, heterogeneous real-time system design. Comput. J. 38 (1995) 271–285
[45] H. Gonz′alez-V′elez and M. Cole. Towards fully adaptive pipeline parallelism for heterogeneous distributed environments. In ISPA 2006, number 4330 in Lect. Notes Comput.Sc., pages 916–926, Sorrento, Dec. 2006. Springer-Verlag.
[46] Gonzalez-Velez, H.; Cole, M., "An adaptive parallel pipeline pattern for grids," Parallel and Distributed Processing, 2008. IPDPS 2008. IEEE International Symposium on , vol., no., pp.1-11, 14-18 April 2008
[47] L.V.Kale and S.Krishnan. Charm++: A portable concurrent object oriented system based on C++. In Proceedings of OOPSLA-93.,March 1993.(Also:Technical Report UIUCDCS-R-93-1796,March 1993,University of Illinois, Urbana, IL.
[48] Kale L.V. and Ramkumar B. and Sinha A.B and Gursoy A. The CHARM Parallel Programming Language and System: Part I - Description of Language Features. IEEE Transactions on Parallel and Distributed Systems, 1994.
[49] The Cilk Project. http://supertech.csail.mit.edu/cilk/
[50] Cilk: An Efficient Multithreaded Runtime System by Robert D. Blumofe, Christopher F. Joerg, Bradley C. Kuszmaul, Charles E. Leiserson, Keith H. Randall, and Yuli Zhou. Proceedings of the Fifth ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP), pp. 207–216, 1995.
[51] Threading Building Blocks Open Source Project.http://www.threadingbuildingblocks.org/
[52] Reinders, James (2007, July). Intel Threading Building Blocks: Outfitting C++ for Multi-core Processor Parallelism (Paperback) Sebastopol: O'Reilly Media, ISBN 978-0-596-51480-8.
[53] Voss, M. (2006, October). "Demystify Scalable Parallelism with Intel Threading Building Blocks' Generic Parallel Algorithms."
[54]李晓明,闫宏飞,王继民.搜索引擎-原理、技术与系统.科学出版社.2004
[55]叶庆华肖利民祝明发,机群作业管理系统的评价体系,南京大学学报(哲学·人文科学·社会科学),2001年3月
[56] Allan Vermeulen, Gabe Beged-Dov, Patrick Tompsom.The Pipeline Design Pattern. Rogue Wave Software,Inc.1995.
[57] Richard H.Carver, Kuo-Chung Tai. Modern Multithreading. John Wiley & Sons,Inc,2006.
[58] Robert D.Blumofe and Charles E.Leiserson. Scheduling multithreaded computations by work stealing. In Proceedings of the 35th Annual Symposium on Foundations of Computer Science. pages 356-268, Santa Fe, New Mexico, November 1994.