全球化的趨勢下，企業組織規模日趨龐大。但對大型組織而言，企業內分支機構之間的企業資料通訊若採行了不合適的方式，便可能無法確保運作順遂，進而造成企業經營的可能損失。
企業內部資料通訊需要發佈更新的資料包含了訊息與檔案。傳統上，企業內部訊息傳遞多採行推送式(push-based)方式進行，如：電話告知、公文傳遞、電子郵件通知等由發送端主動送出訊息的方式。但是這類方式隨著接收端數目增加，容易造成無效率或訊息漏失的情形。而目前大型企業內部資訊系統多為主從式(client/server)架構情況下，必須要經常更新各客戶端應用軟體或資料檔案以維持系統一致性，否則可能因系統無法運作造成企業極大的損失。以往為了維持主從式系統的系統一致性，多採用拉取式(pull-based)的系統更新方式，讓所有的客戶端機器得以從單一伺服器上擷取檔案。若接收端運作正常，此法可以確保資料順利擷取，但這樣的方式，可能造成網路擁塞，也易因伺服器毀損、傳輸量過程中檔案封包擷取錯誤或是網路斷線等導致檔案無法傳遞，進而造成服務中斷與企業損失。
因此，對於企業內部訊息傳遞與系統更新，勢必要進行特別設計，以便能正確且快速傳遞訊息或檔案。過往研究中顯示，推送式資料擷取方式若無完善的容錯機制搭配，無法確保檔案順利送達目的地。而拉取式的方式，只要接收端運作正常，是可以確保接收端順利從發送端擷取資料，但還是可能造成網路壅塞與伺服器負擔過重的情況。因此，故本論文針對應用於企業內部網路之一種可靠與有效的資料發佈機制進行研究；在分析了拉取式與推送式的特點之後，針對訊息傳遞與軟體更新等兩種企業內部資料通訊機制分別進行兼顧效能與容錯的設計。
就訊息傳遞而言，本研究針對經常發佈的訊息，以及吻合企業內部網路的複雜與儘量減少網路流量的要求，設計了整合拉取與推送方式的訊息傳遞機制。透過此機制，得以減少網路頻寬資源耗用。就軟體更新而言，本研究針對企業內部主從式系統的更新進行可容錯的改良式拉取設計，以減少對單一伺服器的依賴，並能成功傳遞檔案。
最後，就訊息傳遞而言，本研究進行投資成本與效用的比較與訊息傳遞等候時間之分析，以供企業建置此類系統之參考。和前一系統比較，本設計更得到了至少82.2%的滿意成效。就軟體更新而言，透過本研究設計之可容錯(fault tolerance)與錯誤移轉(failover)的軟體更新機制，可以減少純粹拉取式可能造成伺服器負擔及避免因災難發生造成系統無更新進而不能提供服務。本研究亦推導此防災難設計之可靠度，以供企業系統更新管理之參考。而在實務運用上，本檔案傳遞機制更協助企業建置了達到了超過六個標準差可靠度標準的軟體更新系統。

Message broadcasting and file delivering are major concerns of a manager to transfer strategic information and keep system consistence within an enterprise. Traditionally, message broadcast uses a push-based scheme to deliver messages from headquarters to front-line staff within a large-scaled company. However, such push-based application could cause message loss. In addition, with industry evolved in fast pace, the software functions are constantly updating, then the changes have to be reflected in all client software through a pull-based file delivery approach before new services can be launched. However, the damage for a large-scaled service company would be caused by an inconsistent system, in which client software fails to receive updated content causing loss of emergency calls and possibility of recovery.
In addition, to avoid losing contents due to the push-based method, companies could adopt pull-based algorithms to deliver contents. However, although the pull-based method can ensure that content could be received, it has a critical problem, the network is easy to be congested. The push-based method can avoid congesting the network, but it needs a specific robust design to ensure that contents could be delivered to destination.
Utilizing pull-based mechanism could resolve problems related to push-based mechanism, but consuming heavy bandwidth. Hence, adopting only a pull-based or a push-based broadcasting algorithm is no longer feasible for a large-scaled company to broadcast messages with different network bandwidth between each connected peer. Therefore, to ensure that every receiver will read downward messages thereby reducing the consumption of network bandwidth, this work proposes a robust push- and pull-based broadcasting system for sending downward messages.
A reliable pull-based content-delivering service is needed for a large-scaled service company to ensure system consistency whenever a disaster occurs. This study also proposes a file delivery mechanism and reveals its availability through analyzing its fault tolerance and failover mechanisms while errors occurred. This work proposes a reliability model of this proposed disaster avoidance mechanism. This disaster avoidance mechanism is designed for preventing damage from disasters, such as earthquakes, tsunamis, nuclear plant explosions or war.
Finally, above proposed message broadcasting mechanism and file delivering mechanism were successfully applied to a large-scaled company. The user satisfaction level of the message broadcasting system was found to be above 82.2% for the five most important factors for measuring user satisfaction. Additionally, empirical results demonstrate that the file delivering mechanism runs on 19 servers with less than 3.4 content delivery failures per million hours is better than a six-sigma capable process.

[1]. Agrawal, M., Rao, H. R., Sanders, G. L.: Impact of mobile computing terminals in police work. Journal of Organizational Computing and Electronic Commerce, 13 (2003) 73-89.
[2]. Abdinnour-Helm, S.F., Chaparro, B.S., Farmer S.M.: Using the end-user computing satisfaction (EUCS) instrument to measure satisfaction with a web site, Decision Sciences, 36 (2005) 341-364.
[3]. Androutsellis-Theotokis, S., Spinellis, D.A.: Survey of peer-to-peer content distribution technologies, ACM Computing Surveys, 36 (2004) 335-371.
[4]. Baily, J. E., Pearson, S.: Development of a tool for measuring and analyzing computer user satisfaction. Management Science, 29 (1983) 530-545.
[5]. Bank, D.R., Telecomm disaster recovery planning for electric utilities. IEEE conference Rural Electric Power 2005: D3/1-D3/10.
[6]. Bartoli, A.: Implementing a replicated service with group communication, Journal of Systems Architecture, 50 (2004) 493-519.
[7]. Birman, K.P., Renesse, R.: Reliable distributed computing with the Isis Toolkit. Wiley: IEEE Computer Society Press, 1994.
[8]. Blake, M. B., Fado, D. H., Mack, G. A.: A publish and subscribe collaboration architecture for web based information. ACM conference WWW (2005) 1164-1165.
[9]. Bhide, M., Deolasee, P., Katkar, A., Panchbudhe, A., Ramamritham, K., Shenoy, P.: Adaptive push-pull: disseminating dynamic web data. IEEE Transactions on Computers, 51 (2002) 652-668.
[10]. Castro, M., Liskov, B.: Practical Byzantine Fault Tolerance and Proactive Recovery. ACM Transactions on Computer Systems, 20 (2002) 398-461.
[11]. Chandra, T. D., Toueg, S.: Unreliable failure detectors for reliable distributed systems. Journal of the ACM, 43 (1996) 225-267.
[12]. Conti, M., Gregori, E., Lapenna, W.: Client-side content delivery policies in replicated web services: parallel access versus single server approach. Performance Evaluation, 59 (2005) 137-157.
[13]. Davies, D. W.: An historical study of the beginnings of packet switching. Computer Journal, 44 (2001) 152.
[14]. Défago, X., Schiper, A., and Urbán, P.: Total order broadcast and multicast algorithms; Taxonomy and Survey. ACM Computing Surveys, 36 (2004) 372-421.
[15]. Denton, D. K.: INTRANETS: Bringing Integration to Performance Measurement. Performance Improvement, 44 (2005) 28-33.
[16]. Fallara, P.: Disaster recovery planning. IEEE Potentials, 22 (2003) 42-44.
[17]. Fich, F., Rparentt, E.: Hundreds of impossibility results for distributed computing. Distributed Computing, 16 (2003) 121-163.
[18]. Fry, M., MacLarty, G.: Policy-based content delivery: an active network approach. Computer Communications, 24(2001) 241-248.
[19]. Goh, T.N.: A strategic assessment of six sigma. Quality and Reliability Engineering International , 18 (2002) 403-410.
[20]. Greenberg, J., Baron, R. A.: Behavior in Organizations, Allyn and Bacon Inc, MA (1993).
[21]. Gunther, N.J.: The dynamics of performance collapse in large-scaled networks and computers. Journal of High Performance Computing Applications, 14 (2000) 367-372.
[22]. Hayes, P.E., Hammons, A., Picking up the pieces: Utilizing disaster recovery project management to improve readiness and response time, IEEE Industrial Application Magazine, 8(2002) 27-36.
[23]. Herrería-Alonso, S., Suárez-González, A., Fernández-Veiga, M., Rubio, R. F. R., López-García, C.: Improving aggregate flow control in differentiated services networks. Computer Networks, 44 (2004) 499-512.
[24]. Hiles, A.: Surviving a computer disaster. Computing & Control Engineering Journal, 3(1992) 271-274.
[25]. Hsieh, C.C., Hsieh, YC.: Reliability and cost optimization in distributed computing systems. Computers and Operations Research, 30 (2003) 1103-1119.
[26]. Jiang, Y., Wu, M.Y., Shu, W.: A hierarchical overlay multicast network. IEEE conference Multimedia and Expo, (2004) 1047-1050.
[27]. Kamilla, K., Håkan, L., Lundberg, L., Svahnberg, C.: Optimal recovery schemes in fault tolerant distributed computing, Acta Inform, 41(2005) 341-365.
[28]. Katsaros, D., Manolopoulos, Y.: Broadcast program generation for webcasting, Data & Knowledge Engineering, 49 (2004) 1-21.
[29]. King, D.: Post Disaster Surveys: experience and methodology, Australian Journal of Emergency Management, 17(2002) 1-13.
[30]. Kleinrock, L.: Queueing Systems Volume I: Theory. John Wiley and Sons, 1975.
[31]. Legris, P., Ingham, J., Collerette, P., Why do people use information technology? A critical review of the technology acceptance model, Information and Management, 40 (2003) 191-204.
[32]. Liang, C.C., Hsu, P.Y., Leu, J.D., Luh, H.: An effective approach for content delivery in an evolving intranet environment- a case study of the largest telecom company in Taiwan. Lecture Notes in Computer Science, 3806 (2005) 740- 749.
[33]. Liang, C.C., Wang, C.H., Luh, H., Hsu, P.Y.: A robust web-based approach for broadcasting down-ward Messages in a large-scaled company. Lecture Notes in Computer Science, 4255 (2005) 222- 233.
[34]. Liang, C.C., Wang, C.H., Luh, H., Hsu, P.Y.: Disaster Avoidance Mechanism for Content-Delivering Service. Computers & Operations Research (Accepted, 5/16/2007).
[35]. Liang, C.C., Wang, C.H., Luh, H., Hsu, P.Y.: Reliability Measure in Content-Distribution Systems. J Computer Science and Network Security, 7 (2007) 138-143.
[36]. Lim, H., Kim, W., Lee, B., Yim, C.: High-speed IP address lookup using balanced multi-way trees. Computer Communications, 29 (2006) 1927-1935.
[37]. Little, J.D.: A Proof of the Queueing Formula , Operations Research, 9 (1961) 383-387.
[38]. Morgan, G., Shrivastava, S.K.: Implementing flexible object group invocation in networked systems. International conference on Dependable Systems and Networks, (2000) 439-48.
[39]. Mills, R.J., Paper, D., Lawless, K.A., Kulikowich, J.M.: Hypertext navigation- an intrinsic component of the corporate intranet. Journal of Computer Information Systems, 43(2002) 44-50.
[40]. Nunnally, J., Bernstein, I.: Psychometric theory, McGraw-Hill, New York, 1994.
[41]. O''Connor, P.D.T.: Commentary: reliability-past, present, and future. IEEE Transactions on Reliability, 49(2000) 335-341.
[42]. Pallis, G., Vakali, A., Insight and perspectives for content delivery networks. Communications of the ACM, 49 (2006) 101-106.
[43]. Pascal, C.L.: Enabling chance interaction through Instant Messaging. IEEE Transactions on Professional Communication, 46(2003) 138-141.
[44]. Pidgeon, N., O’Leary, M., Man-made disasters: why technology and organizations (sometimes) fail, Safety Science, 34(2000) 15-30.
[45]. Ranganathan, C., Ganpathy, S.: Key dimensions of business-to-customer web sites. Information and Management, 39(2005) 457-465.
[46]. Rodeh, O., Birman, K.P., Dolev, D.: Using AVL trees for fault-tolerant group key management, International Journal of Information Security, 1 (2002) 84-99.
[47]. Saxena, N., Pinotti, C. M., Das, S. K.: A Probabilistic push-pull hybrid scheduling algorithm for asymmetric wireless environment. IEEE conference GLOBALCOM (2004) 5-9.
[48]. Shao, B.B.M., Optimal redundancy allocation for information technology disaster recovery in the network economy, IEEE Transactions on Dependable and Secure Computing, 2(2005) 262-67.
[49]. Smith, D.R., Cybrowski, W.J., Zawislan, F., Amstein, D., Dayton, AD, Studwell, TD.: Disaster recovery planning for transmission systems of the defense information system network. IEEE Journal on Selected Areas in Communications, 12(1994) 13-22.
[50]. Stuckenschmidt, H., van Harmelen, F.: Generating and managing metadata for web-based information systems. Knowledge-Based Systems, 17(2004) 201-206.
[51]. Taylor, H.M., Karlin, S.: An Introduction to Stochastic Modeling, Third Edition, Academic Press, 1998.
[52]. Wang, W.M., Liang C.C., Lu H.Z., Chow W.S., Chang K.Y.: Research of testing process: the case of TOPS system delivery process. TL Technical Journal, 34(2004) 7-34.
[53]. Yen, L.H., Chi, K.H.: Maintaining a ring structure for mobile ad hoc computing, Journal of Parallel and Distributed Computing, 64 (2004) 1371-1379.