UranOS: a Cloud-based Resource Management Framework for High Dimensionality Search Optimization

Multidisciplinary automated design-optimization (MDO) is a rapidly growing field of interest in sectors as diverse as architecture, engineering, science, and more. Often the computational cost of doing such optimization can be staggeringly high, requiring hundreds of computers connected through high-performance networks with complex custom designed software to support it. To address these problems, we propose to design, implement and deploy a novel Cloud resource management framework, UranOS, intended for use as a scalable and high resource efficiency resource management platform for MDO applications. UranOS is a lightweight and portable virtualization solution including algorithms, high level languages and tools for performance modeling and efficient resource allocation. Our platform will automatically provide customers with performance optimizations towards improving end-to-end Quality of Service compliance for MDO applications. Also, our industry partner will further benefit from lower operational costs as a result of high resource usage efficiency.

Faculty Supervisor:

Dr. Cristina Amza

Student:

Ali Baradaran Hashemi

Partner:

Autodesk Canada Co

Discipline:

Engineering - computer / electrical

Sector:

Information and communications technologies

University:

University of Toronto