Cloud computing has become a major resource for fulfilling people's computational and storage needs. Investing in these services requires measuring and assuring its quality in general, and reliability and usability are primary concerns. However, using traditional reliability models can be challenging because of the environmental constraints and limited data availability due to the heterogeneous environment and diverse stakeholders. Also, the quality of cloud service Application Programming Interfaces (APIs) has a direct impact on the usability and reliability of the service.

We developed a framework to measure reliability with alternative available information that most cloud providers offer in three stages: 1) Defects are extracted and weighed from issue reports based on their validity, 2) Workload is measured by the number of clients as a new proxy to estimate daily clients usage, 3) Both results are linked together to examine the defect behavior over time. Software reliability growth models (SRGMs) are used to analyze this behavior, to assess current reliability, and to predict future reliability.

Google Maps APIs is used as a case study to demonstrate the applicability and effectiveness of our new framework. Then our framework is validated by extending the models to provide reasonably accurate long term reliability predictions.

Furthermore, we developed a comprehensive framework to measure and analyze cloud service APIs quality attributes in general and usability sub-attributes in particular. First, we identify relevant quality attributes applicable to cloud service APIs. Second, we decompose cloud service APIs to measurable elements. Then we define metrics to quantify these quality attributes using decomposed elements. Lastly, we measure and analyze cloud service APIs usability using existing data sources from crowd source Q&A. We applied our framework on YouTube APIs and Stack Overflow to demonstrate its applicability and effectiveness.

Degree Date

Summer 8-7-2018

Document Type


Degree Name



Computer Science and Engineering


Jeff Tian

Subject Area

Computer Science



Creative Commons License

Creative Commons Attribution-Noncommercial 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License