Deriving the greatest possible benefit from biological systems data poses one of the most significant challenges of the modern era. KU researchers are using advanced computational methods and other information technologies to transform data into useful information. To foster the development and applications of this needed technology, researchers built a computing infrastructure aimed at creating the next generation of algorithms and information integration methodologies required to continue advances in life sciences.
A federal grant from Health Resources and Services Administration helped fund the infrastructure housed on West Campus at the Information and Telecommunication Technology Center (ITTC). The new cluster, made up of 128 processors and 10 servers, permits researchers to fragment computations and perform them in parallel. It is a significant enabler of compute- and data-intensive research at KU. The tightly coupled cluster architecture allows efficient exchange of data between processors.
"The multiple machines housed in one location with dedicated communication between processors on a high-speed network makes it a powerful, integrated platform for researchers," said Terry Clark, assistant professor of EECS, whose teaching and research focus on parallel computing. Clark's students were among the first to use the cluster
The new infrastructure serves capably as a data warehouse with approximately 25 terabytes of storage and a petabyte-capacity backup system, allowing for robust and effective management of the data. The large local storage is needed for data generated from diverse projects such as those involving protein and DNA sequences and high volumes of mass spectrometry data, for example
"Increased storage in combination with faster processing will facilitate turning new data into knowledge of biological systems," said Victor Frost, ITTC director and Dan F. Servey distinguished professor of EECS.