Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.

Author ORCID Identifier



Open Access Dissertation

Document Type


Degree Name

Doctor of Philosophy (PhD)

Degree Program

Electrical and Computer Engineering

Year Degree Awarded


Month Degree Awarded


First Advisor

Michael Zink

Second Advisor

Lixin Gao

Third Advisor

David Irwin

Fourth Advisor

Ramesh Sitaraman

Subject Categories

OS and Networks


As shown in recent studies, video streaming is by far the biggest category of backbone Internet traffic in the US. As a measure to reduce the cost of highly over-provisioned physical infrastructures while remaining the quality of video services, many streaming service providers started to use cloud services, where physical resources can be dynamically allocated based on current demand. In this dissertation, we seek to evaluate and improve the performance for both Adaptive Bitrate (ABR) video streaming and cloud applications. First, we present a set of measurement studies for ABR streaming applications. Using the data from the application, network, and physical layers in different network environments, we identify the key factors that can impact the quality of video delivering services. Then we develop and evaluate a set of new ABR streaming quality adaptation algorithms to improve the user playback experience. In addition, we explore the options for better energy efficiency of ABR video transcoding services and parallel cloud applications. We define a set of energy management policies to enable the utilization of renewable energy sources. We show that, for both applications, by effectively utilizing the renewable energy, our policies can significantly reduce the grid energy usage and corresponding energy cost, while ensuring a satisfying application performance.