FLowell: Accelerating Data-Driven Scientific Research at the University of Massachusetts Lowell

Introduction

We propose to establish FLowell, an enhanced network infrastructure with science DMZ and Software Defined Networking to support a variety of data-driven scientific research at the University of Massachusetts Lowell (UML). Leveraging the existing computing resources, upgrading network equipment and connections, and bringing state-of-the-art technologies, the proposed network infrastructure will be built on computing cluster racks, OpenFlow switches and controllers to enable fine-grained network resource control and sharing, and respond in real time to emerging network needs of researchers in atmospheric science, database science, biomedical science and engineering, computer networking and architecture, representing the science and engineering research activities at UML. FLowell is to be built upon our experience gained through a successfully initial deployment and testing of OpenFlow switching on UMass Lowell campus, in collaboration with Extreme Networks and BigSwitch Networks. Collaborating with researchers beyond campus boundary, FLowell leverages current and new CI resources to accelerate data centric knowledge discovery in a global context.

Specific tasks of this project include: (1) constructing the science DMZ and 10Gbps science
network infrastructure with computing racks and SDN network hardware and software components; (2) developing SDN controller applications to seamlessly control network resource allocation
for the specific needs of the campus research community; (3) participating at-scale data driven science projects using FLowell with collaborating organizations and (4) teaming science researchers and campus IT staff to advance campus cyberinfrastructure through regular task force activities. 

Team

Dr. Yan Luo, PI, Professor, Dept of Electrical and Computer Engineering, University of Massachusetts Lowell

Dr. Ivan Galkin, Co-PI, Research Associate Professor, University of Massachusetts Lowell

Dr. Vinod Vokkarane, Co-PI, Professor, Dept of Electrical and Computer Engineering, University of Massachusetts Lowell

Dr. Tingjian Ge, Co-PI, Associate Professor, Dept of Computer Science, University of Massachusetts Lowell

Dr. Yu Cao, Co-PI, Associate Professor, Dept of Computer Science, University of Massachusetts Lowell

Project Status

Undergoing. Aug 2014 – 2017

Project Sponsor

National Science Foundation, CC*IIE Networking Infrastructure Program

Deliverables

  1. Cao, Yu and Steffey, Shawn and He, Jianbiao and Xiao, Degui and Tao, Cui and Chen, Ping and Müller, Henning. "Medical Image Retrieval: A Multi-modal Approach," Journal of Cancer Informatics (Accepted, impact factor 1.64), 2015.

  2. Hu, Sanqing and Jia, X. and Zhang, J. and Kong, W. and Cao, Yu and Kozma, R.. "Comparison Analysis: Granger Causality and New Causality, and their applications to Motor Imagery (Accepted)," IEEE Transactions on Neural Networks and Learning Systems (TNNLS) (Impact Factor: 4.37), 2015.

  3. Huang, Xiaobing and Zhao, Tian and Cao, Yu and Brown, C.. "Pipeline Information Retrieval: A Domain Specific Language for Multimedia Information Retrieval," International Journal of Multimedia Data Engineering and Management (IJMDEM) (Impact factor: 0.98), v.5, 2014, p. 1-27.

  4. Chen Tian, Jingdong Sun, Weimin Wu, Yan Luo. "Optimal Bandwidth Allocation for Hybrid Video-on-Demand Streaming with a Distributed Max Flow Algorithm," Computer Networks, v.91, 2015, p. 483. doi:http://dx.doi.org/10.1016/j.comnet.2015.08.049 

  5. Chen Xu, Xiaoban Wu, Yan Luo, Brian Tierney and Jeronimo Bezerra, Pepple: Programmable Network Measurement for Troubleshooting Soft Failures, IEEE 37th Sarnoff Symposium, Newark NJ, Sept 19-21, 2016.

  6. Lu He, Tim Miskell, Rui Liu, Hengyong Yu, Huijuan Xu, Yan Luo. "Scalable 2K K-SVD Parallel Algorithm for Dictionary Learning on GPU," ACM International Conference on Computing Frontiers 2016, 2016.

  7. Peilong Li and Yan Luo. "P4GPU: Acceleration of Programmable Data Plane Using a CPU-GPU Heterogeneous Architecture," IEEE 17th International Conference on High Performance Switching and Routing, 2016.