一個以雲端運算進行系統效能測試之系統

A System Performance Testing System Based on Cloud Computing

張耀庭(Yao-Ting Chang)

指導教授 ： 鄭有進

國立臺北科技大學/機電學院/資訊工程系所/碩士(2012年)

摘要

當企業推出軟體服務時，如果伺服器預期會服務大量使用者，則必須額外確認軟體服務伺服器的系統效能需求，以確保軟體服務的品質與穩定性。一般在執行系統效能測試時，為了模擬大量的使用者，必須用到大量的資源與設備，並且會耗費相當多的人力於部署測試環境與蒐集數據。本論文提出一套以雲端運算進行的系統效能測試系統，透過雲端基礎設施服務提供虛擬機器，並藉由雲端運算的可擴性，有效利用虛擬機器的資源。測試人員執行測試前，先提供測試腳本與待測伺服器，再透過系統自動執行測試流程，最後產生系統效能數據報表。企業透過分析測試數據找出伺服器在提供服務時，系統效能負載的瓶頸，作為伺服器調整或擴充之依據。本論文以Apache WebDAV Server作為測試服務進行實驗，實驗顯示使用雲端虛擬機器存取待測服務，伺服器系統效能測試結果與使用實體機器相近。

關鍵字

系統效能測試 ； 雲端基礎設施服務

並列摘要

When an enterprise launches a new software service designed for a large number of users, it is important to evaluate the system performance of the server on which the service runs, so that quality of the software service can be ensured. A system performance evaluation normally needs to simulate large number of users, which requires many computer equipments (called test machines) and is both resource-intensive and labor-intensive. This thesis, proposes a system performance testing system integrated with the cloud, which provides test machines on infrastructure as a service (IaaS). Therefore, the number of test machines can be scaled easily. Before executing a system performance testing with the proposed system, the tester needs to setup a server under test and prepare a test script. Our system can automatically perform the testing and generates a test report. The bottleneck of the software service (or the server) can be identified by analyzing the test report. An experiment based on the proposed system has been conducted for the system performance testing of the Apache WebDAV Server. The results indicated that test machines on the cloud can replace physical machines, providing similar system performance testing results.

並列關鍵字

System Performance Testing ； Infrastructure as a Service

參考文獻

[1] Peter Mell, Timothy Grance, “The NIST Definition of Cloud Computing,” NIST Special Publication 800-145.

[4] James C. King, “Symbolic execution and Program Testing,” Communications of the ACM, 1976.

[7] Devarshi Ghoshal, R.Shane Canon, Lavanya Ramarkrishnan, “I/O Performance of Virtualized Cloud Environments,” DataCloud-SC’11.

[2] Lian Yu, Wei-Tek Tsai, Xiangji Chen, Linqing Liu, Yan Zhao, Liangjie Tang and Wei Zhao, “Testing as a Service over Cloud,” 2010 Fifth IEEE International Symposium on Service Oriented System Engineering.

[3] Lian Yu, Le Zhang, Huiru Xiang, Yu Su, Wei Zhao, Jun Zhu, “A Framework of Testing as a Service,” Proceedings of the Conference of Information System Management, 2009.