臺灣博碩士論文加值系統

隨著硬體及網路科技的進步，工作站群集(workstation cluster)逐漸成為平行及分散式處理重要的一環，與傳統中央式超級電腦比較，工作站群集的優點在於花費少量金錢而提供相當的運算能力，再配合網路的連結，工作站群集更可成為低價位的平行電腦。本論文JPDCE(Java Parallel Distributed Computing Environment)這個系統就有這樣的特性，系統中整合了廉價的個人電腦和工作站群來組成一部虛擬的平行電腦；整個系統主要由管理系統(系統管理者、節點管理者、工作管理者、控制台)和檔案系統等所組成，負責計算資源的收集、程式的分派、執行等複雜的過程，這讓使用者只需專注於程式的設計，當程式開發測試完成，再交與其本人桌面上的工作機器，最後執行結果將展現於其面前，這對平行分散式處理的程式設計師而言是個很方便的系統。管理系統和檔案系統所組成的JPDCE是整合異質性的環境，因此我們選擇以Java來製作系統，因為Java有跨平台的特性可以解決異質環境共通性的問題，也提供了物件導向程式環境，這對程式的開發維護及使用有很大的助益。

With the advancement of hardware and network technology, workstation clusters become an important role in parallel and distributed processing. The workstation clusters are cheaper than traditional super computers. With the network connection, the workstation clusters will be a low price parallel computer. Our system Java Parallel Distributed Computing Environment (JPDCE) is a heterogeneous environment and is a virtual parallel computer. We use Java to implement the system. Java can solve the problem of heterogeneous environment and support object oriented. Object oriented program is easy to reuse and maintain. JPDCE is a complete heterogeneous parallel and distributed environment. JPDCE consists of system manager, node manager, job manager, console and file system. The user just focus on coding program don’t care the collection of resource, dispatch the program and execution. Those jobs are done by our virtual machine. The usr just summit the program to our system then waiting the result. This environment is very convenient for programmer, and it is also the character of our system.

第一章 簡介…………………………………………………….1
第二章 相關研究…………………………………………………….5
第三章 系統架構…………………………………………………….9
第一節 系統特色…………………………………………………….10
第二節 管理系統…………………………………………………….11
第三節 檔案系統…………………………………………………….12
第四節 JPVM…………………………………………………….13
第五節 工作執行流程…………………………………………………….13
第四章 管理系統…………………………………………………….16
第一節 系統管理者…………………………………………………….17
第二節 節點管理者…………………………………………………….19
第三節 工作管理者…………………………………………………….21
第四節 控制台…………………………………………………….22
第五章 檔案系統…………………………………………………….25
第一節 動機…………………………………………………….25
第二節 WebNFS…………………………………………………….26
第三節 檔案系統架構…………………………………………………….27
第六章 實驗結果…………………………………………………….32
第一節 實驗環境…………………………………………………….32
第二節 矩陣運算…………………………………………………….33
第三節 圓週率運算…………………………………………………….34
第四節 質數的運算…………………………………………………..36
第七章 結論…………………………………………………….39
參考文獻…………………………………………………….40

[1] Baratloo, A., Karaul, M., Kedem, X., and Wyckoff, P. Charlotte: Metacomputing on the Web. In Proc. 9th Int''l Conf. on Parallel and Distributed Computing Systems (Sept. 1996).
[2] Beberg, A., Lawson, J., and McNett, D. distributed.net. http://www.distributed.net, 1997.
[3] Beguelin, A., Dongarra, J. J., Geist, G. A., Manchek, R., and Moore, K. HeNCE: a heterogeneous network computing environment. Scientic Programming 3, 1 (Spring 1994), 49-60.
[4] Berman, F., Wolski, R., Figueira, S., Schopf, J., and Shao, G. Application- Level Scheduling on Distributed Heterogeneous Networks. In Proc. of 1996 Int''l Conf. on Supercomputing (1996).
[5] Bricker, A., Litzkow, M., and Livny, M. Condor Technical Summary. Tech. Rep. 1069, Dept. of Computer Science, University of Wisconsin-Madison, 1992.
[6] Callaghan, B. WebNFS. Whitepaper, Sun Microsystems, Inc., Apr. 1997.
[7] Chandy, K. M., Rifkin, A., Sivilotti, P., Mandelson, J., Richardson, M., Tanaka, W., and Leisman, L. A World-Wide Distributed System Using Java and the Internet. In Int''l Symp. on High Performance Distributed Computing (1996), IEEE.
[8] Devaney, J. E., Lipman, R., Lo, M., Mitchell, W. F., Edwards, M., and Clark, C. W. The Parallel Applications Development Environment (PADE) User''s Manual. Techrep., National Institute of Standards and Technology, Nov. 1995.
[9] Distributed.net. Press Release. http://www.distributed.net/pressroom/56-pr.html, Oct. 1997.
[10] Fagg, G. E., Moore, K., Dongarra, J. J., and Geist, A. Scalable Network Information Processing Environment (SNIPE). In Super Computing (1997).
[11] Feinefeld, A., Baraglia, R., Decker, T., Gehring, J., and el. al. The MOL Project: An Open, Extensible Metacomputer. In Proc. 6th Heterogeneous Computing Workshop (1997), pp. 17-31.
[12] Ferrari, A. J. JPVM: Network Parallel Computing in Java. Tech. Rep. CS-97-29, Dept. of Computer Science, Univ. of Virginia, 1997.
[13] Foster, I., and Kesselman, C. Globus: A Metacomputing Infrastructure Toolkit. International Journal of Supercomputer Applications (1997), 115-128.
[14] Hayes, B. Collective Wisdom. American Scientist 86, 2 (1998), 118-122.
[15] Judd, G., Clement, M., and Snell, Q. Dogma: Distributed Object Group Man- agement Architecture. Tech. Rep. BYU-NCL-97-102, Computer Science Dept., righam Young Univ., 1997.
[16] Litzkow, M. J., Livny, M., and Mutka, M. W. Condor: A Hunter of Idle Workstations. In Proc. of the 8th Int''l Conf. on Distributed Computing System (ICDCS-8) (1988),pp. 104-111.
[17] Neuman, B. C., and Rao, S. The Prospero Resource Manager: A Scalable Framework for Processor Allocation in Distributed Systems. Concurrency: Practice and Experience 6, 4 (June 1994), 339-355.
[18] Philippsen, M., and Zenger, M. JavaParty - Transparent Remote Objects in Java. Currency: Practice and Experience 9, 11 (1997), 1225-1242.
[19] Regev, O., and Nisan, N. The popcorn marcket - an online markets for computational resources. In 1st Conf. on Information and Computation Exonomics (1998).
[20] Smarr, L., and Catlett, C. E. Metacomputing. Communication of the ACM 35, 6 (June 1992), 45-52.
[21] Sullivan, III, W. T., Werthimer, D., Bowyer, S., Cobb, J., Gedye, D., and Anderson, D. A New Major SETI Project Based on Project Serendip Data and 100,000 Personal Computers. In Proc. 5th Intl. Conf. on Bioastronomy (1997).
[22] Woltman, G. Great internet mersenne prime search. http://www.mersenne.org/, 1998.