" /> " /> NSF Awards Investigators over $3.25M for Advanced Networking Research | UMass Lowell Skip to Main Content

NSF Awards Investigators over $3.25M for Advanced Networking Research

Projects Will Develop Technologies, Accelerate 'Big Data' and Facilitate Biomedical Studies

High-speed computer networking
University researchers are building a campus network cyberinfrastructure that would enable them to access and share vast datasets faster and more efficiently.

By Edwin L. Aguirre

We live in the age of “Big Data.” Every second, massive amounts of digital information are produced, stored, retrieved, analyzed and transferred around the world from such diverse sources as hospitals, military and defense agencies, revenue and census agencies, social networks, Internet search engines, stock markets, airlines, insurance and credit card companies.

To harness the power of Big Data, an interdisciplinary team of university researchers are building a campus network cyberinfrastructure that would enable investigators to quickly access and share vast datasets with their counterparts at the Amherst, Boston, Dartmouth and Worcester campuses as well as other external collaborators and industry partners. It will also allow for faster and more efficient virtual experiments, simulations and modeling through the Massachusetts Green High Performance Computing Center in Holyoke.

That effort recently received a boost from the National Science Foundation (NSF) when the agency awarded more than $3.25 million to four projects that support high-speed computer networking research at UMass Lowell.

“These high-impact projects will position the university in a leading role in advancing network technologies and accelerating Big Data and biomedical research,” says Assoc. Prof. Yan Luo of the Department of Electrical and Computer Engineering (ECE). “The methodologies developed from the project can ultimately lead to breakthroughs or discoveries in a wide range of endeavors, from studying the weather and monitoring the environment to fighting terrorism, designing a new drug and finding cures for deadly diseases.”

Enhancing Data Flow with ‘FLowell’

The first project — called “FLowell” — will accelerate data-driven science projects at UMass Lowell by enhancing the campus cyberinfrastructure with software-defined network (SDN) technologies to speed up data transmission rate tenfold, to 10 Gbps (billion bits per second) via fiber-optic cables. Construction of the infrastructure, which is supported by a two-year, $512,000 NSF grant, is expected to be completed by the end of summer.
Assoc. Prof. Yan Luo with student in the lab
Assoc. Prof. Yan Luo, right, and a graduate student work in the Advanced Computing and Networking Systems Lab.

“The university is seeing a tremendous growth of data-intensive research in many disciplines, such as atmospheric science, materials science, biomedical science and data visualization,” notes Luo, who is the project’s principal investigator (PI). “Right now, firewalls slow down the data stream, causing bottlenecks in the network. Our new SDN-based system will use the dedicated science Demilitarized Zone, or DMZ, concept, which will eliminate bottlenecks when transferring large datasets.”

In addition to Luo, the project’s co-PIs are ECE Assoc. Prof. Vinod Vokkarane, computer science Asst. Prof. Yu Cao and Assoc. Prof. Tingjian Ge, and Assoc. Director Ivan Galkin of the university’s Space Science Laboratory.

The university’s Office of Information Technology also played a key role in this campus-wide collaboration. “We are excited to be partners in the project and are eager to see how this effort would improve the flow of science data across the university’s network,” says Michael Cipriano, the chief information officer for the university.

Adds Ralph Zottola, chief technology officer for research computing in the UMass President’s Office: “It is exciting to see that the investment made by the UMass system in the Massachusetts Green High Performance Computing Center to support our researchers is paying dividends so early and with such important work.”  

Measuring Network Traffic and Usage

The second project — called the Advanced Measurement Instrument and Services (AMIS) — will allow researchers to measure data-flow levels at the connection points of ultrahigh-speed networks to understand network usage and traffic patterns, identify and troubleshoot network anomalies and provide insights to network control and planning.

“AMIS will equip network operators and administrators with tools to accurately monitor and quantify the network’s flow-level performance and conduct in-depth traffic analysis while preserving user privacy,” says Luo.

Luo is the project’s PI, with UMass Lowell as the lead institution. Other academic collaborators include UMass Boston, the University of Kentucky and the University of Texas, El Paso. UMass Lowell’s share of the funding is nearly $1,255,000, which is spread over three years.

Biomedical Informatics Innovation

The third project focuses on the development of a robust and scalable network infrastructure for biomedical informatics research.
Patient with blood-pressure monitor
Researchers are developing a high-performance cyberinfrastructure for transferring biomedical sensor data collected from patients to medical facilities and/or data analytics platforms.

“We are building a high-performance cyberinfrastructure specifically to support efficient transfer of large quantities of heterogeneous biomedical sensor data collected from patients to medical facilities and/or data analytics platforms,” says Vokkarane. “Applications of this technology can include real-time control of instruments and transport of real-time video and other time-sensitive data.”

The NSF funding for this project is $1 million over two years, with Vokkarane as PI and Luo and Cao as co-PIs. Other collaborators include researchers from UMass Medical School in Worcester as well as physicians from the University of Tennessee’s College of Medicine in Chattanooga.

“The effort will have significant impacts on improving the quality of health-care applications, providing clinical and scientific researchers with flexible and efficient network resource allocation for studying patient behavior and for training the next generation of workforce in medical and engineering fields,” notes Vokkarane.

As an example, the technology is being applied in rural Peru to help physicians and medical practitioners improve the screening and diagnosis of local residents afflicted with tuberculosis using smartphones connected to cloud-computing servers.

Securing Patients’ Personal Health Information

The goal of the final project — called Secure Transport and Research Architecture for Monitoring Stroke Recovery (STREAMS) — is to assist with the care and rehabilitation of stroke patients at home. This is accomplished by making the live data streams generated by the patient’s monitoring devices (wearable sensors and webcams) secure as the data are transmitted via the home’s Wi-Fi network to cloud-computing data centers for real-time processing and analysis. 

Nurse with stroke patient
The goal of the STREAMS project is to assist with the care and rehabilitation of stroke patients at home.
The NSF awarded nearly $500,000 over two years for the project, with Luo as PI and Cao, computer science Assoc. Prof. Xinwen Fu and ECE Prof. Martin Margala as co-PIs. Other collaborators include researchers from UMass Medical School and the University of Tennessee in Chattanooga.

Sensor-generated live data streams of the patients’ status and activities can help family members, caregivers or healthcare professionals to respond to the patients’ needs quickly and effectively. “For example, the data streams can help predict the patients’ risk of accidental falls,” says Luo. “They can also tell people if the patients have been taking their medications on time and are meeting their daily dietary requirements.”

He adds: “Since the data streams contain personal health information, they must be protected using advanced encryption/decryption algorithms.”

The four projects will help the NSF to quantify the network usage for Big Data research projects and serve as a model for other high-speed networks to follow, says Luo.