不管在工作方面或是在休閒方面，網際網路服務對人們來說都變得日益重要。熱門的網際網路服務往往在無法提供服務的時候造成巨大的經濟損失。由於相當長的執行時間及數量龐大的客戶端連線，網際網路服務常常遭受不可預期的錯誤或是軟體年紀過老的問題。重新啟動服務可以解決上述產生的問題但是將網際網路服務整個重新啟動需要花費相當長的一段時間。因此，重新啟動服務會造成網際網路服務無法提供服務的時間過長。
有許多已經被提出的研究可以有效得縮短服務重新啟動所花費的時間。在這些研究之中，網際網路服務可以達到更高的可利用度但是卻會遭受到效能上的損耗，而上述的狀況導因於這些研究之中的網際網路架構特性。
我們提出了一個架構，此架構包含了在64位元Linux作業系統上的單位址空間延伸架構及此架構對幫助網際網路服務在不損耗效能的情況下達到高可利用度的支援。在這個我們提出的架構之中，我們有效的利用了逐漸普及的64位元位址空間。必須達到高可利用度的網際網路服務會在經過修改之後在我們的單位址空間延伸架構下執行，而傳統應用程式則仍舊在傳統的多重位址空間環境下執行。

Internet service has become very important no matter in business or in our leisure time. Popular Internet service always results in great economical lose while service crash. Due to the long execution time and the huge number of clients, Internet services can easily suffer from transient faults or the software aging problem. Service restart can solve these problems but fully restart of an Internet service may take a long period of time. Thus fully restart will cause the Internet service downtime too long.
Some researches have been proposed to reduce the service restart time efficiently. Internet services may achieve higher availability in these researches but will suffer from performance degradation because of the framework characteristic in these researches.
We proposed a framework which includes a Single-Address-Space extension on 64-bit Linux and supports to help the Internet service achieve higher availability without much performance degradation as previous researches. In the framework, we make good use of the widespread 64-bit address space. The Internet service with the need of higher availability will be modified to run efficiently in the Single-Address-Space extension and the legacy applications can still run in the traditional Multiple-Address-Space environment.

[1] An analysis of Web Server performance with an accelerator//proof
[2] CoW: A Cooperative Multithreading Web Server//proof
[3] Performance Technologies Inc.,“The effects of network downtime on profits and productivity - a white paper analysis on the importance of non-stop networking”, available at http://whitepapers.informationweek.com/detail/RES/991044232_ 762.html, 2001.//proof
[4] D. Oppenheimer, D. A. Patterson, “Why do Internet services fail, and what can be done about it?”, In Proceedings of the 10th ACM SIGOPS European Workshop, Saint-Emilion, France, Sep. 2002.//proof
[5] Measurement, Analysis and Performance Improvement of the Apache Web Server//proof
Single-Address-Space
[6] Uwe Dannowski, Andreas Haeberlen, Gernot Heiser, Espen Skoglund, Volkmar Uhlig, “Performance of Address-Space Multiplexing on the Pentium”, Technical Report 2002-1, Department of Computer Science, University of Karlsruhe, May 2002. //sas
[7] Improved Address-Space Switching on Pentium Processors by Transparently Multiplexing User Address Spaces//sas
[8] Intra-Address Space Protection Using Segmentation Hardware//sas
[9] The Opal Single-Address-Space Operating System//sas
[10] The Mungi Single-Address-Space Operating System//sas
[11] Porting Linux to x86/64//x86/64
[12] AMD64 Architecture Programmer’s Manual Volume 1: Application Programming
[13] AMD64 Architecture Programmer’s Manual Volume 2: System Programming
[14] AMD64 Architecture Programmer’s Manual Volume 3: General-Purpose and System Instructions//x86/64

Fault Tolerant
[15] D. Patterson, A. Brown, P. Broadwell, G. Candea, M. Chen, J. Cutler, P. Enriquez, A. Fox, M. Merzbacher, D. Oppenheimer, N. Sastry, W. Tetzlaff, J. Traupman, N. Treuhaft, "Recovery-Oriented Computing (ROC): motivation, definition, techniques, and case studies", Computer Science Technical Report UCB//CSD-02-1175, Mar. 2002.//component
[16] G. Candea, A. Fox, “Crash-Only Software”, Proc. 9th Workshop on Hot Topics in Operating Systems, pp.67-72, Jun. 2003//recovery
[17] G. Candea, J. Cutler, A. Fox, “Improving Availability with Recursive Microreboots: A Soft-State System Case Study”, Performance Evaluation Journal, Vol. 56, No. 1-3, Mar. 2004
[18] Y. Chawathe, E. A. Brewer, “System Support for Scalable and Fault Tolerant Internet Service”, Proc. IFIP International Conference on Distributed Systems Platforms and Open Distributed Processing (Middleware ’98), Sep. 1998.//component
[19] Flexible and Reliable Web Service Execution
[20] Recovering Internet Service Sessions from Operating System Failures//recovery
[21] Detecting Application-Level Failures in Component-based Internet Services//fault detection
[22] Implementation and Evaluation of Transparent Fault-Tolerant Web Service with Kernel-Level Support//recovery
[23] Apache Modeling Project