(3.238.235.155) 您好!臺灣時間:2021/05/16 06:49
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

: 
twitterline
研究生:吳奕翰
研究生(外文):I-HAN
論文名稱:提供長程資源可用度預測以進行計算網格之工作排程
論文名稱(外文):Supporting Long-Term Resource-Availability Prediction for Job Scheduling in Computational Grid
指導教授:梁廷宇
指導教授(外文):Tyng-Yeu Liang
學位類別:碩士
校院名稱:國立高雄應用科技大學
系所名稱:電機工程系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2009
畢業學年度:97
語文別:中文
論文頁數:73
中文關鍵詞:網格計算資源監控資源選擇工作排程資源可用度
外文關鍵詞:Grid computingResource monitoringResource selectionJob schedulingResource availability
相關次數:
  • 被引用被引用:0
  • 點閱點閱:254
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
摘 要

網格技術的發展主要為了提供了強大的計算能力資源。然而在網格計算環境中,所有共享的計算資源是隷屬於不同的資源擁有者,故因此這些計算資源兼具有動態性以及非專屬性。此兩種性質會讓系統很難確實掌握資源運行的狀態以及未來資源負載的情況,因此可能在資源配置上做出錯誤的決定,進而影響使用者程式的執行效能。雖然在過去許多關於資源監控與資源選擇的文獻被提出,但大多數研究的貢獻只限於資源的監控,無法兼備預測的功能。在資源選擇時,往往也只是考慮看當下的資源負載狀態或者基於短時間的資源效能預測。但是網格的工作所執行的時間通常是需要很長一段時間,倘若資源的配置只是考慮當下或者未來短時間內的資源負載勢必是不夠的。為了解決這些問題,在本論文成功發展出一套資源監控系統並且可提供長程資源可用度的預測,以幫助進行網格計算之工作排程,並且能夠為工作配置到一組適合資源,有效地提升工作執行的效率。
Abstract

The development of Grid technology offers a powerful computation capability in scientific, academic, business, etc. However, the computational resources provided by the Grid are dynamic and non-dedicated because the resources usually belong to diverse to owners. The future execution states and the load conditions of the resources can hardly be controlled by the due to these to characteristics. Sometimes they may cause error decisions on resource allocations and therefore the performance of the user’s applications. There are many studies and researches in the fields of resource monitoring and resource selection proposed in the past. Most of these studies focused on the resource monitoring function but not on the prediction capability of the resource management system. When the system selects resources to match the job requests, it considers the current state of workload or sometimes based on the prediction of short-term availability of the resources. However, job execution on the Grid system usually take relatively long period of time, it is not enough if the resource selection considered only the current state or the short-term resource availability. This thesis proposed and developed a resource monitoring system that can provide long-term prediction of resource availability for job scheduling in order to allocate proper sets of resource for job execution in a long period of time as well as increase the efficiency of job execution.
目 錄
目 錄 iv
圖目錄 vi
表目錄 vii
第1章 概論(Introduction) - 1 -
1.1研究動機(Research Motivation) - 1 -
1.2研究目的(Research Purpose) - 3 -
1.3論文架構(Thesis Organization) - 3 -
第2章 相關文獻與研究背景(Related work and Background) - 4 -
2.1監控資源負載系統(Monitor Resource information system) - 4 -
2.1.1 Ganglia分散式監控資源系統 - 4 -
2.1.2 Network Weather service(NWS)網路流量監控服務 - 5 -
2.2負載預測(Workload Predictions) - 7 -
2.2.1中央處理器負載預測(CPU Workload Predictions) - 7 -
2.2.2網路傳輸速度預測 (Network Bandwidth Predictions) - 9 -
2.3背景知識(Background) - 13 -
第3章 資源監控系統 - 17 -
3.1資源監控系統設計考量 - 17 -
3.2資源監控系統架構組織 - 19 -
3.3循式樣式探勘預測方式(Sequential Pattern Mining Forecast Method) - 21 -
3.3.1負載等級的劃分與負載樣式表格的定義 - 21 -
3.3.2未來CPU可用度的預測 - 23 -
3.3.3未來Bandwidth可用度的預測 - 24 -
3.3.4未來Memory可用度的預測 - 24 -
第4章 系統實作(System Implementation) - 26 -
4.1資源監控系統架構組織構成元件 - 26 -
4.2資源監視器(Resource Monitor) - 29 -
4.3資源感測器(Resource Sensor) - 31 -
4.3.1資源負載樣式的收集 - 31 -
4.3.2資源負載樣式的探勘 - 34 -
4.3.3資源可用度的預測 - 37 -
第5章 效能評估(Performance Evaluation) - 40 -
5.1實驗環境與方式 - 40 -
5.1.1預測準確度的評估方式 - 40 -
5.1.2資源選擇效能的評估方式 - 41 -
5.2資料探勘代價(Mining cost) - 47 -
5.3預測準確度的評估結果 - 50 -
5.4資源選擇效能的評估結果 - 54 -
第6章 結論與未來工作(Conclusions and Future work) - 60 -
參考文獻 (References) - 61 -
參考文獻 (References)
[1]A. Schuster, R. Wolff, D. Trock, “A High-Performance Distribute Algorithm for Mining Association Rules”, knowledge and Information Systems, Page 458-475, 2005

[2]C.H Wang, H.K. Huang, H.L. Li, “A Fast Distribute Mining Algorithm for Association Rules with Item Constraints”,IEEE International Conference on Systems, Man, and Cybernetics, Page 1900-1905, 2000.

[3]D.W. Cheung, J.W. Han, V.T. Ng, A.W. Fu, “A Fast Distributed Algorithm for Mining Association Rules”, The 4th International Conference on Parallel and Distributed Information Systems, Page 31-42, 1996.

[4]Gaidioz, B., Wolski, R., and Tourancheau, B. “ Synchronizing Network Probes to avoid Measurement Intrusiveness with the Network Weather Service” , The 9th IEEE High-performance Distributed Computing Conference, Page 147-154, 2000,.

[5]I. Foster, C. Kesselman, “The Grid2: Blueprint for a New Computing Infrastructure”, Morgan Kaufmann Publishers, 2004.

[6]I. Foster, C. Kesselman, “The Anatomy of The Grid: Enabling Scalable Virtual Organizations”, International Journal of High Performance Computing Applications, Page 200-222, 2001.

[7]I. Foster, C. Kesselman, J.M. Nick, S. Tuecke, “Grid Services for Distributed System Integration”, IEEE Computer, Page 37-46, 2002.

[8]J.J. Lu, B.W. Xu, L. Xu, D.Z. Kang, H.W. Chenm, H.J. Yang, “Mining Association Rules with Linguistic Terms”, Proceeding of the 15th International Conference on Tools with Artificial Intelligence (ICTAI’03), Page 129, 2003.

[9]K. Czajkowski, S. Fitzgerald, I. Foster, C. Kesselman. “Grid Information Services for Distributed Resource Sharing” , Proceedings of the Tenth IEEE International Symposium on High-Performance Distributed Computing (HPDC-10), IEEE Press, 2001.

[10]K. Hu, D.W. Cheung, S.W. Xia, “Efficient Parallel Mining of Association Rules on Shared-Memory Multiple-Processor Machine”, 1997 IEEE International Conference on Intelligent Processing Systems, Page 1133-1137, 1997.

[11]L. Yang, I. Foster, and J.M. Schopf , “Homeostatic and Tendency-based CPU load Predictions”, International Symposium on Parallel and Distributed Processing (IPDPS'03), 2003.

[12]M.L. Massie, B.N. Chun, D.E. Culler, “The Ganglia Distributed Monitoring System: Design, Implementation, and Experience”, Parallel Computing, Page 817-840, 2004

[13]M. Baker, R. Buyya, D. Laforenza,“Grids and Grid Technologies for Wide-area Distributed Computing”, Software: Practice and Experience , Pages 1437-1466, 2002.

[14]M. Cannataro, D. Talia, P. Trunfio, “Distributed Data Mining on the Grid ”, Future Generation Computer Systems, Pages 1101-1112, October 2002.

[15]O.B. Yaik, C.H. Yong, F. Haron, “Time Series Prediction Using Adaptive Association Rules”, First International Conference on Distributed Frameworks for Multimedia Applications (DFMA'05), Page 310-314, 2005.

[16]R. Chu, N. Xiao and X. Lu, “A Resource Information Management System for RAM Grid”, International Conference on Grid and Cooperative Computing (GCC 2007), 2007.

[17]R.Wolski, N.T. Spring, J. Hayes “The Network Weather Service A Distributed Resource Performance Forecasting Service for Metacomputing”, Journal of Future Generation Computing Systems, Page 757-768,1999

[18]R. Wolski, N. Spring, and J. Hayes, “Predicting the CPU Availability of Time-shared Unix Systems,” Proceedings of 8th IEEE High Performance Distributed Computing Conference (HPDC 1999), Page 102-115, 1999.

[19]R. Agrawal and R. Srikant, “Mining Sequential Patterns”, IEEE International Conference on Data Engineering, Page 3-14, 1995.


[20]R. Agrawal, T. Imielinski, A. N. Swami, “Mining Association Rules between Sets of Items in Large Data Bases”, Proceedings of the 1993 ACM-SIGMOD International Conference on Management of Data, Page 207-216, 1993.

[21]R. Agrawal, J.C. Shafer,“Parallel Mining of Association Rules”, IEEE Transactions on Knowledge and Data Engineering, , Page 962-969, 1996.

[22]R. Wolski,“Dynamically Forecasting Network Performance Using the Network Weather Service” ,Journal of Cluster Computing, 1998.

[23]S. Fitzgerald, I. Foster, C. Kesselman, G. von Laszewski, W. Smith, S. Tuecke. “A Directory Service for Configuring High-Performance Distributed Computations” IEEE Symposium on High-Performance Distributed Computing, Page 365-375, 1997.

[24]R.M Wolski, N. Spring, C. Peterson, “ Implementing a Performance Forecasting System for Metacomputing: The Network Weather Service” , Proceedings of SC97, November, 1997.

[25]W. Yongwei,Y. Yulai, Y. Guangwen, Z. Weimin,“Load Prediction Using Hybrid Model for Computational Grid”, Proceedings of The 8th IEEE/ACM International Conference on Grid Computing, Page 235-242, 2007.

[26]Y. Yulai, .Yongwei, Y.Guangwen, Z. Weimin “Adaptive Hybrid Model for Long Term Load Prediction in Computational Grid”, Proceedings of The 8thIEEE International Symposium on Cluster Computing and the Grid (CCGrid 2008), Page 340-347.

[27]Y. Zhang, W. Sun, Y. Inoguchi, “CPU Load Predictions on the Computational Grid”, Sixth IEEE International Symposium on Cluster Computing and the Grid(CCGRID’06), Page 321-326, 2006.


[28] 張良毅, “發掘負載樣式以支援計算網格之資源選擇”, 碩士論文, 國立高雄應用科技大學, 2006
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊