You are here

Engineering researchers collaborate on NSF cloud computing testbed

Computer Science and Engineering Professor Dhabaleswar K. (DK) Panda and his team will play a key role in a $10 million cloud computing research project announced recently by the National Science Foundation (NSF).

Prof. DK PandaCo-located at the University of Chicago and The University of Texas at Austin, the NSFCloud testbed called Chameleon will support the design, deployment and initial operation of a large-scale, reconfigurable experimental environment for cloud research. Partners on the Chameleon project include The Ohio State University, Northwestern University and the University of Texas at San Antonio.

Cloud computing refers to the practice of using a network of remote servers to store, manage and process data, rather than a local server or a personal computer. In recent years, cloud computing has become the dominant method of providing computing infrastructure for Internet services.

Chameleon will consist of 650 cloud nodes with 5 petabytes of storage. Researchers will be able to configure slices of Chameleon as custom clouds using pre-defined or custom software to test the efficiency and usability of different cloud architectures on a range of problems, from machine learning and adaptive operating systems to climate simulations and flood prediction.

Ohio State’s Network-Based Computing Laboratory, led by Prof. Panda, will focus on high performance network protocols and middleware critical to cloud storage, processing and sharing.

Panda’s team designed and developed MVAPICH, a high-performance message passing interface (MPI) library for clusters with InfiniBand, iWARP and RoCE interconnects. The MPI software is currently being used by more than 2,200 organizations in 73 countries, empowering many TOP500 systems around the world.

According to Panda, in the Chameleon project, MVAPICH as well as other big data middleware from his group will be enhanced to work in a physical server—also known as bare metal—cloud environment. The aim will be to deliver high-performance and scalability together with virtualization.

"Delivering bare metal performance for big compute research and big data to complement virtualization has been an open-challenge in the community for many years,” he said. "Our team will try to achieve this and provide new ways to design, deploy and use next generation cloud environments."

For more details, read the full NSF release.