Quantitative X-ray imaging facility is moving big bytes across the network

The teeth and jaw fossil fragments from a potential new human ancestor species that roamed regions of Ethiopia more than 3 million years ago are just a few of the more noteworthy objects Tim Ryan has scanned at Penn State's Center for Quantitative X-ray Imaging (CQI). The fossils might add a new branch to humanity’s family tree and provide further evidence that multiple human ancestors once roamed Africa alongside one another.

“We’ve scanned many fossils from locations abroad, but this was a one-of-a-kind specimen specifically brought to Penn State to be scanned at our facility,” said Ryan. “These types of fossils rarely leave their museums, let alone their countries. So it was exciting for me to conduct the scan and compile data to share with other researchers so we can make new discoveries about our early human ancestors.”

Ryan, an associate professor of anthropology and information sciences and technology, was scanning the prehistoric fossils to create high-resolution, 3-D digital models that paleoanthropologists could analyze to further prove the existence of the new human ancestor species without damaging or destroying the original, rare fossil specimens.

The fossil discovery of another possible human ancestor has important implications for our understanding of early human ecology and also raises significant questions as to how prehistoric humans living at the same time and in the same geographic area might have shared the landscape and available resources.

This kind of high-impact research is perfect for the CQI, where, as co-director, Ryan helps facilitate the use of micro-computed tomography or “micro-CT” imaging technology.

The CQI is one of fewer than 15 academic facilities worldwide that provides shared access to large-scale academic, industrial X-ray micro-CT imaging. The center has also recently installed a new, state-of-the-art CT system that is able to scan in minute detail. Since the new scanner produces data sets much larger than the previous scanner, the center is one of several research areas at the University that stores their data on the Institute for CyberScience Advanced CyberInfrastructure (ICS-ACI) — an advanced computing and storage system that supports research — and leverages the Penn State ‘Big Data’ Research Network to transfer data.

In addition to prehistoric jawbones, the CQI facility has also scanned such varied objects as rocks to study the impact of fracking, mouse heads to better understand skull deformities in children and M&Ms to study the uniformity of the candy coating.

The size of all this data is large and can run up to 10 and 20 gigabits for a single data set. However, most studies generate multiple data sets that are stored in many individual files within a larger folder, making it hard to manage. As the data is further manipulated by researchers, it can get even bigger.

Storing and transferring such large volumes of data can be daunting. In the past, the center has traditionally used software utilities that transfer computer files from one computer to another over the shared Penn State network or Internet. The center has also downloaded the data on CDs or jump drives.

“While these methods work fine when the data is in the range of a few megabits, it doesn’t work very well when you have 10s, 100s or even 1000s of gigabits of data,” said Ryan.

More often than not, researchers would just download the data into smaller chunks onto their desktop. “Then they’d delete that chunk of data and download the next chunk and repeat until they were done,” said Ryan.
 
Although the CQI used to scan and store its data on the College of Earth and Mineral Sciences high-speed network, transfer outside of the college's network was limited by the gigabit connection to the University network, so the whole transfer process was lengthy.

Now, by connecting to the Penn State ‘Big Data’ Research Network, the process is quicker and more efficient because the research network is optimized for high-performance scientific applications (rather than for shared, general-purpose administrative systems or academic computing).

“The Penn State Research Network enables the CQI to do file transfers without such bottlenecks as academic firewalls, varying connection speeds or competition with non-research traffic, making the data transfer process much more efficient,” said Tom Canich, IT manager in the College of Earth and Mineral Sciences. As a result, Canich says the new process is up to 10 times faster than the previous option.

In addition to an increase in transfer speed, storing the data at the ICS-ACI will also enable other analysis and visualization research activities to be performed on scans done at the CQI, making it even easier for researchers like Ryan to continue to help make new discoveries.

For more IT stories at Penn State, visit http://news.it.psu.edu.

Contacts: 
Last Updated February 25, 2016