Author Type

Graduate Student

Date of Award

Summer 7-28-2026

Document Type

Thesis

Publication Status

Version of Record

Submission Date

August 2026

Department

Computer and Electrical Engineering and Computer Science

College Granting Degree

College of Engineering and Computer Science

Department Granting Degree

Electrical Engineering and Computer Science

Degree Name

Master of Science (MS)

Thesis/Dissertation Advisor [Chair]

Borko Furht

Abstract

Cloud computing depends on load balancing to allocate user requests among virtual machines (VMs), affecting response time, utilization, and scalability. Round Robin, Equally Spread Current Execution, and standard Throttled scheduling are widely used; however, standard Throttled selects among eligible VMs using an implementation-dependent order with no explicit preference. This thesis proposes the Weighted Throttled Load Balancing (WTLB) algorithm, which introduces an explicit, deterministic selection priority among eligible VMs, implemented by modifying the Throttled scheduler within CloudAnalyst.

Evaluated under an identical six-region, multi-data-center configuration against Round Robin, ESCE, and standard Throttled, WTLB preserves the aggregate response-time, processing-time, and cost profile of standard Throttled while changing its internal VM-selection rule. All four algorithms produced the same aggregate response time, processing time, and cost at the reported precision. WTLB’s contribution is therefore its deterministic, configured-priority selection mechanism and its preservation of macro-level performance, rather than a reduction in response time.

Download

Share

COinS