網格計算之通訊安全設計

Design of Secure Communication for Grid Computing

程瑞芳(Jui-Fang Cheng)

指導教授 ： 詹昭文

國立臺中科技大學/資訊與流通學院/資訊工程系碩士班/碩士(2014年)

https://doi.org/10.6826/NUTC.2014.00063

摘要

資訊社會發達，人們大量使用資訊技術，來完成部份計算量低的工作。資訊革命下一步就是要協助人們使用資訊技術來完成計算量高的工作。由於，計算量高的工作需由高階的電腦完成，而高階的電腦價格較高，使用者不易獨自擁有，且計算量高的工作非經常性的工作，而使用者獨自擁有高階電腦也不合理。近年來，學者提出網格計算的概念，來協助人們完成計算量高的工作。網格計算利用成熟的網路通訊與資源共享技術，讓一個群組的中低階價位之電腦，組成相當於一個虛擬高階電腦，來提供人們計算量高的服務。這些組成網格計算的中低階電腦分屬不同的擁有者。這些電腦在閒置的狀況下，即可加入網格計算提供服務，但是，擁有者欲使用他的電腦時，該電腦應與一個事先定義好的群組電腦，協商出一台閒置的電腦來取代它在網格計算中的工作。若所執行之高計算量工作是機密性的，則上述協商過程也應受安全之保護。本文主要研究，一台對擁有者而言，由閒置狀態轉變為忙碌狀態之電腦，使電腦須與事先定義的群組電腦，協商出一個群組秘密金鑰，與挑出一台閒置的電腦來取代它在網格計算中的工作。近年來，大部份的網格計算研究，在執行群組成員的金鑰協定時都要透過驗證中心(CAS)，而這將會增加驗證中心的負擔。基於二次剩餘，本研究提出一個低運算量和傳輸量的群組成員通訊安全協定。在傳輸過程中除了可以保護資料，而且資料不會被他人存取或修改。雖然群組通訊成員增加，但是傳送端與接收端的計算量卻不會有明顯的差異。當群組金鑰協定在執行的時候，它不需要經過驗證中心。則驗證中心即可以降低負擔和提升資訊服務的品質。我們為未來的研究提供安全與高效能的運算平台，可加速重要研究之進行。

關鍵字

二次剩餘(Quadratic Residue) ； 信賴驗證中心(CAS) ； 通訊安全 ； 網格計算

並列摘要

With the development of information society, information technology is largely utilized for completing part of the tasks which require low computation complexity. The next step of information revolution is to facilitate people to complete the tasks which require high computation complexity. High computation complexity tasks must be completed by high-end computers. Nevertheless, as the price of a high-end computer is comparatively high, the user could hardly possess one alone. Usually, high computation complexity tasks are not regular tasks. It’s also unreasonable for the user to have a high-end computer all alone. In recently years, the concept of grid computing has been addressed by scholars to assist people in completing high computation complexity tasks. Grid computing enables a group of low-medium priced computers to compose a virtual high-end computer for providing people high computation complexity services using mature network communication and resources sharing technology. The low-medium end computers which compose grid computing are held by various owners. As these computers are under idle condition, they could join in grid computing and provide services. Nonetheless, when a owner intends to use his computer, the computer must make a negotiation with the computer group defined in advance to have an idle computer replace its grip computing task. Provided the proceeding high computation complexity task is confidential, then, the forgoing negotiation process must be under security protection, too. Accordingly, the main purpose of this research is as follows: Regarding the computer of a owner which has converted from an idle state to the busy state, the computer must negotiate with a computer group defined beforehand to generate a secret group key and select an idle computer from it for replacing the task in grid computing. Over the recent years, in the research on grid computing, mostly, in executing the key protocol of group members, it’s required to go through CAS (Central Authority System), which would add burden to CAS. Based on quadratic residue, in this research, a secure communication protocol for group members with low computation complexity and transmission quantity was proposed. In the transmission process, aside from data protection, it could also forbid data from being accessed or altered by others. Although the number of group communication members increases, there would be no significant differences in the computation complexity at the sender and receiver end. In executing the group key protocol, it’s unnecessary to pass through CAS. This could, then, reduce the burden of CAS and enhance the information service quality. We, hereby, provide a secure and high-effect computation platform for further research to facilitate the proceeding of significant research.

並列關鍵字

Central Authority System (CAS) ； Grid computing ； Quadratic residue ； Secure communication

參考文獻

[3] J. Masque, A. Peinado, Cryptanalysis of improved Liaw’s broadcasting cryptosystem, Journal of Information Science and Engineering, Vol. 22, 2006, pp. 391–399.

[1] W. Chung, R. Chang, “A new mechanism for resource monitoring in grid computing”, Future Generation Computer Systems, Vol. 25, No. 1, 2009, pp. 1–7.

[2] M. Smith, M. Schmidt, N. Fallenbeck, T. Dornemann, C. Schridde, B. Freisleben, “Secure on-demand grid computing”, Future Generation Computer Systems Vol.25, No. 3, 2009, pp. 315–325.

[4] X. Zou, Y. Dai, X. Ran, “Dual-level key management for secure grid communication in dynamic and hierarchical groups”, Future Generation Computer Systems, Vol. 23, No. 6, 2007, 7pp. 76–786.

[5] H. Sun, “Security of broadcasting cryptosystem in computer networks”, Electronics Letters, Vol. 35, 1999, pp. 2108–2109.