Mechanisms and Algorithms for Optimal Use of Inter-Clouds

Cloud computing is one of the pillars of the modern computing infrastructure mainly because it allows the procurement of computing services on a pay-as-you-go basis. However, despite the many benefits offered by cloud computing, it has several significant drawbacks such as data lock-in, lack of universal geographic proximity, risk of service outages, and variable cost structures. To address these problems, researchers from the academia and industry have proposed an interconnection of many independently managed cloud-computing systems such that workloads can be seamlessly migrated among them to suite the customer requirements. Many research problems are yet to be solved to realize the vision of interconnecting cloud-computing systems. The goal of this project is to develop new lightweight architectures for interconnecting cloud-computing systems. In particular, we will focus on the development of a Linux container-based workload-mapping framework that will benchmark the different clouds, select the appropriate clouds, and migrate workloads to maximize client benefits. We want to realize the inter-cloud architecture without requiring the clouds to agree on standardized cloud management protocols.

Faculty Supervisor:

Muthucumaru Maheswaran

Student:

Partner:

CloudOps

Discipline:

Computer science

Sector:

Information and cultural industries

University:

McGill University

Program: