臺灣博碩士論文加值系統

由於需求的增加，使得越來越多的雲端服務出現，但目前來說雲端服務仍有許多進步的空間。我們建議了一個能夠整合公有雲端服務的架構，可以降低成本並增加其彈性。其中我們考慮了三個問題：（1）一般的雲端存儲服務不適合處理大而複雜的數據，然而一個NoSQL的資料庫，是最常見拿來進行大數據處理的數據庫，他能夠穩定的處理許多企業的業務訊息及資料。 （2）若要將多個雲端服務整合成一個集中的系統或應用程序是具有挑戰的。 （3）企業經常使用自己內部的設備，而這些企業所購買的設備往往佔成本的很大一部分。

本研究旨在透過多個免費的雲端NoSQL資料庫來構建跨服務的架構，用來虛擬化及整合所有所需的運算資源及服務。本研究中，我們建立了一個基於所提出架構的股票交易模擬系統，其開發是透過各種基於雲端的服務，來整合及虛擬化運算資源並降低設備成本，並且最終實際測試所提出架構的可行性及性能。

More and more cloud services emerging because of increased demand, but there is significant room for improvement. We propose architecture to integrate public cloud services for reducing costs and increasing their flexibility. We consider three issues: (1) The general cloud storage services are unsuitable for large and complex data. A NoSQL datastore, the most common type of database for big data, can process operational information. (2) Integrating several cloud services into centralized systems or applications is challenging. (3) Enterprises often use their own equipment, which accounts for a large part of their costs.
This study aims to construct cross-service architecture via free cloud-based NoSQL datastores for virtualizing and integrating needed computing resources and services. We build a stock trading game to test the feasibility and performance of the proposed architecture which can reduce equipment costs by virtualizing computing resources through various cloud-based services.

1. Introduction 1
2. Research Background 3
2.1 Cloud Services 3
2.2 Cross Services 3
2.3 Service-Oriented Architecture (SOA) 3
2.4 Cloud Service APIs 4
2.5 Big Data 4
3. Architecture & Design 5
3.1 Web-based Application 6
3.2 Cloud-based Application 7
3.2.1 MongoDB 8
3.2.2 Google Cloud Datastore 9
3.2.3 Hadoop Hbase 9
3.2.4 Integrated Cross-Services System 10
3.3 Mobile version Application 12
3.4 Process of Big data 13
3.5 System analysis 15
3.5.1 Functions 15
3.5.2 Diagrams 16
4. Service Features 21
5. Experiment & Evaluation 23
5.1 Mobile applications testing 23
5.2 Web applications testing 25
6. Conclusion & Future Work 30
6.1 Different Access Method 30
6.2 Data Storage Method 31
7. Reference 32

[1]S.E. Chang, K.-M. Chiu , and Y.-C. Chiao, “Cloud Migration: Planning Guidelines and Execution Framework,” Proc. 2015 Seventh Int. Conf. Ubiquitous and Future Networks, pp. 814-819, IEEE, 2015.
[2]G. Breiter and V.K. Naik, “A Framework for Controlling and Managing Hybrid Cloud Service Integration,” 2013 IEEE Int. Conf. Cloud Eng., pp. 217–224, IEEE, 2013.
[3]V. Beal, “Cloud Services,” Webopedia, Retrieved 15 Nov. 2015. Available at: http://www.webopedia.com/TERM/C/cloud_services.html.
[4]C. Xue and Z. Zeng, “The Construction of SOA-Based Enterprise Information Management System,” Proc. 2010 International Conference on Computer Application and System Modeling, vol. 6, pp. 522–525, IEEE, 2010.
[5]F. Lo, “Bigdata Technology－What is Hadoop? What is NoSQL? What is MapReduce?” Retrieved 20 Nov. 2015. Available at https://datajobs.com/what-is-hadoop-and-nosql
[6]MongoDB Inc., “MongoDB for GIANT Ideas | MongoDB.” Retrieved 5 Jan. 2016. Available at https://www.mongodb.com
[7]Google, “Datastore‒NoSQL Schemaless Database, Google Cloud Platform.” Retrieved 13 Oct. 2015. Available at https://cloud.google.com/datastore/
[8]BlazeClan Inc., “Build Your Big Data Building Blocks with AWS Cloud.” Retrieved 5 Jan. 2016. Available at http://www.blazeclan.com/build-your-big-data-building-blocks-with-aws-cloud/
[9]Testdroid Inc., “Testdroid - Mobile App Testing on Android and iOS Devices” Retrieved 20 April. 2016.. Available: at http://testdroid.com/
[10]Pingdom, “Pingdom Website Speed Test.” Retrieved 22 Dec. 2015. Available at http://tools.pingdom.com/fpt/
[11]Load Impact, “On Demand Load Testing for Developers & Testers.” Retrieved 21 Dec. 2015.. Available: at https://loadimpact.com
[12]M. Rouse, “Heartbeat,” TechTarget. Retrieved 22 Dec. 2015. Available at http://searchenterpriselinux.techtarget.com/definition/Heartbeat