The driving force behind our research group, in the Department of Electrical and Computer Engineering and Computer Science at the University of Toronto, is the fascination with complex distributed systems, and the crucial problem of achieving coherent self-organization within such systems.

Led by Dr. Amza, we have successfully built a state-of-the-art self-organizing Cloud infrastructure, called Chorus. Chorus is a networked

Cloud platform providing support for data analytics in Web-based services. For this purpose, Chorus maintains significant amounts of data for various services in a persistent and consistent fashion.

Chorus currently offers more flexible, and customizable Cloud access for users than Amazon; it also offers a set of novel algorithms, languages and performance modeling and diagnosis tools towards cost-effective self-organization. While Chorus has its roots in hosting complex workloads for e-commerce and gaming,  we are going to extend its customization capabilities towards hosting biomedical images and data analytics for them.

The challenge for medical experts is cross-correlating image data of many patients, e.g., brain fMRI’s, because the data is so massive. Our hope is to be able to host large amounts of biomedical data and thus enable medical staff from different communities to share knowledge.  As well, there is a significant opportunity for us to contribute by utilizing our data analytics tools to extract patterns, trends, or uncover abnormalities and semantic correlations in e-health data.