International Journal of Modern Education and Computer Science (IJMECS)
IJMECS Vol. 2, No. 1, 8 Nov. 2010
Cover page and Table of Contents: PDF (size: 279KB)
A High-Performance Communication Service for Parallel Servo Computing
Full Text (PDF, 279KB), PP.9-16
Views: 0 Downloads: 0
Author(s)
Cheng Xin
1,*
Zhou Yunfei
1
Hu Yongbin
1
Kong Xiangbin
2
Index Terms
Communication service, Parallel computing, VME bus, data transmission
Abstract
Complexity of algorithms for the servo control in the multi-dimensional, ultra-precise stage application has made multi-processor parallel computing technology needed. Considering the specific communication requirements in the parallel servo computing, we propose a communication service scheme based on VME bus, which provides high-performance data transmission and precise synchronization trigger support for the processors involved. Communications service is implemented on both standard VME bus and user-defined Internal Bus (IB), and can be redefined online. This paper introduces parallel servo computing architecture and communication service, describes structure and implementation details of each module in the service, and finally provides data transmission model and analysis. Experimental results show that communication services can provide high-speed data transmission with sub-nanosecond-level error of transmission latency, and synchronous trigger with nanosecond-level synchronization error. Moreover, the performance of communication service is not affected by the increasing number of processors.
Cite This Paper
Cheng Xin, Zhou Yunfei, Hu Yongbin, Kong Xiangbin, "A High-Performance Communication Service for Parallel Servo Computing", International Journal of Modern Education and Computer Science(IJMECS), vol.2, no.1, pp.9-16, 2010. DOI:10.5815/ijmecs.2010.01.02
Reference
[1]Fujita Itaru, Sakai Fumio, Uzawa Shigeyuki, Next generation scanner to sub-100 nm lithography, In Proceedings of SPIE. Santa Clara, CA, USA, 2003, pp. 811-821.
[2]Yang Liangliang, Zhou Yunfei, Pan Haihong, Luo Fuyuan. Research on synchronous mechanism of step-scan projection lithography, China mechanical engineering, 2009(20), pp. 20-23, 43.
[3]Wang Chunhong, Hu Jinchun, Zhu Yu, Yin Wensheng, Optimal synchronous trajectory tracking control of wafer and reticle stages. Tsinghua Science and Technology, 2009 (14), pp. 287-292.
[4]Message Passing Interface Forum, MPI: A Message-Passing Interface Standard, Technical Report CS-94-230,Computer Science Department, University of Tennessee, 1994.
[5]J. Bruck, D. Dolev, C. Ho, M. Rosu, R. Strong, Efficient message passing interface (MPI) for parallel computing on clusters of workstations, Journal of Parallel and Distributed Computing, 1997(40), pp. 19- 34.
[6]M. Lauria, A. Chien, MPI-FM: High performance MPI on workstation clusters, Journal of Parallel and Distributed Computing, 1997(40), pp. 4-18.
[7]W. Gropp, E. Lusk, N. Doss, A. Skjellum, A high-performance, portable implementation of the MPI message passing interface standard, Parallel Computing, 1996(22), pp. 89- 828.
[8]R. Hempel, D.W. Walker, The emergence of the MPI message passing standard for parallel computing, Computer Standards & Interfaces, 1999(21), pp. 51- 62.
[9]TMS320C6713B floating-point digital signal processor, http://focus.ti.com.cn/cn/docs/prod/folders/print/tms320c6713b.html
[10]Yan Luxin , Zhang Tianxu , Zhong sheng, Parallel system architecture of multi-DSP interconnected by FPGA, Systems Engineering and Electronics, 2005 (27), pp. 1757-1759, 1775.
[11]James Kohout, Alan D. George, A high-performance communication service for parallel computing on distributed DSP systems, Parallel Computing, 2003(29), pp. 851- 878.
[12]Bai Yucheng, Tang Xiaoqi, Chen Jihong, Hu Huan, Research on multi-axis synchronous control for network CNC system, manufacturing technology & machine tool , 2008 (8), pp. 61-65.
[13]Justin (Gus) Hurwitz, Wu-chun Feng, Analyzing MPI performance over 10-Gigabit Ethernet, Journal of parallel and distributed computing, 2005(65), pp. 1253-1260.
[14]Amit Karwande, XinYuan, DavidK. Lowenthal, An MPI prototype for compiled communication on Ethernet switched clusters, Journal of parallel and distributed computing, 2005(65), pp. 1123-1133.
[15]U.S. VMEbus International Trade Association. American National Standard for Front Panel Data Port Specifications. U.S. Scottsdale. 1998
[16]Zhang Huachun, Sun Changyu, A New Design Method for High-Speed Data Acquisition System Based on FPDP, Systems Engineering and Electronics, 2003(25), pp. 787-789, 803.
[17]Zhao Chunhui, Yang Shuyuan, Yuan Jianping, Design of DSP-specific VME-64 bus interface of NUMA multi-processors systems, Systems Engineering and Electronics 2005 (27), pp. 189-192.