Model-View Resource Allocation in the Cloud

Description

With an ever-increasing number of tenants running workloads in the cloud, a large number of technical challenges arise. For example, how to characterize dynamic workloads from different tenants and how to schedule dynamic workloads in an effective way. In this project, we will analyze resource allocation with model-view approach. Models could capture the inherent correlations (e.g., temporally and spatially) in the data stream through splitting data into disjoint segments and approximating each segment with different types of models (e.g., regression and probabilistic.).

In this project, you will play with Big Data (about 200GB) on resource allocation data from Google clusters. You will find it interesting to approximate the data with different models like regression, probabilistic models. Based on the models you have built, we could apply forecasting query to predict the future resource demands.

• Having the motivation for indulging in a research oriented project
• Programming skills with Java and experience on MapReduce is a plus
• Familiar with basic Machine Learning and Data Mining

this project is already assigned.

Contacts

In case of any questions, please drop us an email or come to our offices:

• Hao Zhuang (BC 118)
• Tian Guo (BC 141)