Deep learning on the cloud

Faculty Mentor: Ying Mao (Fordham)

Virtualization is a promising technology that has facilitated cloud computing to become the next wave of the Internet revolution. Adopted by data centers, millions of applications that are powered by various virtual machines improve the quality of services. Although virtual machines are well-isolated among each other, they suffer from redundant boot volumes and slow provisioning time.

To address limitations, containers were born to deploy and run distributed applications without launching entire virtual machines. However, general-purpose container orchestrion platforms, such as Docker Swarm and Kubernetes, fail to work well with deep learning applications, e.g. Tensorflow and Pytorch. With the respect to unique characteristics, we aim to improve a cluster of containerized deep learning applications and boost the overall system performance.

Learning and Research Objectives

The participants of this project will achieve the following learning and research objectives:

  1. Learn the basic concepts of deep learning applications.
  2. Learn how to understand the python programs that develop with Tensorflow and Pytorch libraries.
  3. Investigate the containerized deep learning applications.
  4. Explore the optimization techniques to improve the performance of the targeted applications.

tensorflow pytorch docker k8s