September 27, 2021

cloud covid article

-Part of our Aggie Innovation Platform series

Two research efforts from the Texas A&M University College of Engineering are using data-driven research methods combined with cloud technology to combat the impact of the COVID-19 pandemic. These projects address different problems across different scales, from local to global.

These crucial research initiatives have one thing in common: they are both some of the first Texas A&M research projects to utilize the Aggie Innovation Platform (AIP) cloud technology.

Raising the Bar — In the Cloud

Dr. Ali Mostafavi, associate professor in the Zachry Department of Civil and Environmental Engineering, and his Urban Resilience.AI Lab began using the cloud for research in 2017. Their predictive AI analysis improves community resilience

His research on prediction and response to urban disaster situations kicked into high gear after Hurricane Harvey. Mostafavi collected data on traffic changes, satellite imagery, social media,and flood impacts — and all of this data needed to be stored securely with sufficient CPUs to process.

“My lab had a server, but not one with enough storage or computing capacity,” states Mostafavi.

Mostafavi applied for and received a grant from AWS in 2017. He used the credits from the award for his hurricane-focused analysis from 2017-20. When the pandemic hit in 2020, Mostafavi’s research focus changed and his data use exploded.

“I went from 20TB to 150TB in just three months. This rapid growth completely consumed my Amazon AWS credits.” 

Mostafavi applied to Microsoft’s AI for Health program for additional funding. Initially receiving $50,000 in credits, followed by an additional $55,000, Mostafavi knew he needed a partner to help him make the best use of his cloud infrastructure.  

College of Engineering IT and the AIP support team helped Mostafavi transition his research data to the AIP. 

“They were tremendously helpful. I didn’t even know what I didn’t know.” says Mostafavi. “AIP’s direct network connection to campus is a game changer. And by partnering with IT, I was able to avoid costly data egress charges.” 

Create a New Collaborative Advanced Computing Solution? No Problem!

Dr. Zenon Medina-Cetina, associate professor and holder of the Zachry Career Development Professorship II in the Department of Civil and Environmental Engineering, is the lead researcher of two projects hosted by Texas A&M’s Cross-Border Threat Screening and Supply Chain Defense (CBTS), a Department of Homeland Security (DHS) Center of Excellence. 

Medina-Cetina’s lab, the Stochastic Geomechanics Laboratory (SGL), assesses the state of risk of complex processes varying in space and time, such as the state of risk for U.S. global supply chains. Currently, the SGL is specifically focused on the impact of COVID-19 on supply chains with Mexico, the United States’ largest trading partner. SGL’s core research relies on the use of spatio-temporal probabilistic methods and algorithms, which demand significant computing power and storage capacity.

A core component of CBTS’ mission is to inform stakeholders of the state of risk of any U.S. supply chain. To this end, Medina-Cetina’s team developed the CBTS-SGL COVID-19 Binational Dashboard, using a new ‘Texas A&M-on-premise and cloud’ advanced computing collaborative model, the CBTS Data-Lake System (CBTS-DLS). This hybrid on-premise/cloud computing model is capable of producing near- to real-time risk analytics describing the evolution of COVID-19, current and emerging threats, vulnerabilities and impacts to health supply chain systems — and the likely impact that a combination of these may cause to society, the economy and the environment.

“In the past, I have always preferred to manage my own computing services on campus. But these COVID projects presented unique technical challenges resulting in the design of a new ‘Texas A&M-on-premise and AWS cloud’ advanced computing collaborative model,” states Medina-Cetina.

“Working with Engineering IT went so smoothly. With both an on-premise and cloud system, we really needed their help to connect the dots and they delivered.”

CBTS now is supported by a set of dedicated servers in Texas A&M’s data center to perform the high-power data crunching. The AIP cloud component serves up this valuable data via their COVID dashboard to the rest of the world.

“With a team of more than 50 international professionals, working together across multiple organizations, all the pieces of our research need to work together properly,” stresses Medina-Cetina. “If the CBTS Data Lake System (CBTS-DLS) doesn’t work properly, nothing else works. It’s the core of everything. Our CBTS-SGL dashboard is the means to communicate our research findings to the world, and it’s hosted in the cloud!”

Benefits of the Cloud

So why are Mostafavi and Medina-Cetina proponents of the cloud in general and the Aggie Innovation Platform in particular? 

“I have confidence when I’m using the cloud I don’t have to worry about security, backups or power,” declares Mostafavi. “I’m confident when we use the cloud our data is safe, it’s protected and we won’t lose it.”

Medina-Cetina followed the advice of its IT team and it paid off. “I trusted my IT team and DHS approved our new technology design immediately — and they take security very seriously.”  

The CBTS also benefits from the straightforward and flexible billing process. “I can transparently show my sponsors exactly where their money is going and how it is being spent,” states Medina-Cetina. “They are very happy with what we are doing and know we are managing the project with the highest standards possible.”

Research in AIP is:

    • Affordable - Billing process is streamlined and tax exemption is taken care of. 
    • Accessible - Data can be accessed by research partners across the world. New partners or grad students can be added quickly and start working immediately.
    • Efficient - Maintenance is much easier no more backups or security patches.
    • Scalable - The cloud is flexible. Scaling back or scaling up resources is simple.
    • Secure - Highly secure systems protect important data with no extra effort.
    • Supported - Local IT staff and the AIP team make complex systems straightforward. 
    • Reliable - Data is always available and accessible. 

Ultimately, being in the cloud reduces risk 

“Security is the number-one feature that won me over,” explains Medina-Cetina. “We are dealing with so many data variables, and while so far these are public, the analytics are very sensitive. The cloud provides a very robust approval and security process. This made us take the risk... and we always favor reducing risk!” 

Future of Big Data Research
Mostafavi and Medina-Cetina agree the benefits of cloud technology make it a perfect partner for data-driven research. 

Data-driven research is growing and big data is the future. The natural solution for this research is the cloud,” says Mostafavi. “The NSF has a cloud bank and funding agencies are already promoting cloud. As a PI, the cloud is great. I don’t have to worry — I can focus on my work and not administrative tasks.” 

Medina-Cetina agrees. “From my experience, our work with the AIP in the design of our hybrid on-premise/cloud platform could become a collaborative model for breaking silos across converging research groups based anywhere in the world.” 

---

The Division of IT has partnered with the top public cloud platforms to provide a flexible, secure and compliant cloud computing experience for the Texas A&M community. The Aggie Innovation Platform (AIP) provides access to these robust services at no added overhead costs, and with the added benefit of a Master Contract, customized networking configuration, and campus subdomain and identity management services.

The AIP has been developed to provide the Texas A&M community with access to a simple-to-start, flexible, scalable, secure, and compliant place to create and innovate with minimal barriers or long-term financial investment risk.

If you want to learn more, visit the AIP website. For questions or to order services, email aip@tamu.edu.