Link to NUIT Home

NUIT HOME

Summer 2014 Edition

eCommunicator is a quarterly newsletter about technology tools, services, and support at Northwestern University.

eCommunicator Home sidebar links

Article Sidebar Links

In this issue…

NUIT Events & Presentations

What’s New & Changing in IT?

eCOMMUNICATOR

Summer 2014 Edition

Northwestern Supports Researchers in Search of Big Data Solutions

Imagine searching for one particular email out of the hundreds you receive in your inbox. Annoying, but achievable. Now imagine the challenge of scrolling through a trillion emails from millions of individual senders to perform that same task. Welcome to big data.

Big data is a phrase commonly used to describe massive volumes of data so large and complex that it’s difficult to process using traditional database and software techniques. Often, getting meaningful information from such large amounts of data is challenging and requires efficient processing power, analytics capabilities, and skills.

Northwestern is committed to providing opportunities to help those at the University who routinely work with big data address these challenges.

Finding Solutions to Big Data Problems

Virtual School session at Northwestern

Last month, Northwestern University partnered with the University of Chicago Research Computing Center (RCC) to host the Virtual School of Computational Science and Engineering (VSCSE) Data Intensive Summer School which focused on the skills needed to manage, process, and gain insight from large amounts of data. View the agenda.

More than 75 Northwestern graduate students, post docs, and staff attended the three-day interactive workshop which was simultaneously delivered to 20 participating sites nationwide using high-definition videoconferencing technology. Attendees were able to participate in hands-on lab activities with on-site support from skilled computational specialists from the RCC and Northwestern University.

“The University’s involvement in the Virtual School offers unique opportunities to enrich the research capabilities of our graduate students, post docs, and staff who are seeking new and effective methods and tools to obtain meaningful results from their research,” said Joseph Paris, associate director of NUIT A&RT Research Computing Services.

Sifting out the Signal from the Noise

Among those representing Northwestern was Kathryn Alpert, a software developer for the Neuroimaging and Applied Computational Anatomy Lab in the Department of Psychiatry and Behavioral Sciences at the Feinberg School of Medicine. She writes code that processes and analyzes neuroimages and provides programming support and instruction for Northwestern graduate students and local high school students working in Linux environments.

Alpert utilizes the parallel computing capabilities of Quest, the University’s high performance computing cluster, to efficiently run image processing algorithms for data manipulation, extraction, and analysis. She then combines these routines into processing pipelines, which run on Quest and are available to colleagues through the Northwestern University Neuroimaging Data Archive (NUNDA), a data archival, management, and processing resource for the Northwestern neuroimaging community.

Alpert says that by attending the virtual summer school she gained a greater awareness of the different big data management solutions and insight on how to deal with various types of big data problems.

“After data processing (which can be a big data problem itself), I often end up with hundreds of thousands data points (independent variables) for each subject, some informative and some uninformative,” said Alpert. “That’s a lot of data! Without the proper computing resources, and a deep understanding of how to approach big data, it is impossible to sift out the signal (helpful stuff) from the noise, or in other words, to isolate the informative information from the junk.”

Got Data Too Big to Handle?

NUIT offers expert assistance to help you find the computational resources—both internal and external to University—for better managing your large research data. Visit the Research Computing Consultation Services web page to schedule a one-on-one consultation and learn about solutions available to approach your big data challenges.

Learn More

The Virtual School of Computational Science and Engineering is a national virtual organization whose goal is to develop and deliver a computational science curriculum that accelerates the ability of faculty, staff, and students to use emerging computational resources to advance science and engineering.

Funding and support for the Virtual School are provided by the Great Lakes Consortium for Petascale Computation (GLCPC), the National Science Foundation (NSF), the State of Illinois, the Committee on Institutional Cooperation (CIC), and Internet2 Commons.

Visit the VSCSE home page for information on summer courses remaining in August and a list of completed and viewable computational science webinars.



Social Media