The High Performance Computing Centre Stuttgart (HLRS) of the University of Stuttgart has deployed a new big data system. It is one of the world’s first data analytics systems of this size in a production environment. HLRS addresses in this way current challenges in high performance computing. With the new system, HLRS enables its customers for the first time to analyze petabytes of data efficiently in real-time.
The new computer, a Cray Urika-GX system, has a total of 64 compute nodes; each equipped with 36-cores of Intel’s current Xeon processor generation. Nodes are staffed with 512 gigabytes of random-access memory, as well as 1.6 terabytes of local SSD memory. Thus, the system is optimally designed to provide solutions for current and future challenges of data analytics. The new data analytics system supports innovative big data technologies such as Hadoop and Spark, both of which boost data analytics in engineering. Moreover, the special-puropose Cray Graph Engine enhances the analysis of semantic data, which is commonly present in biology and chemistry.
The new Urika-GX system was installed at HLRS premises in December 2016. HLRS started at the same time a new big data project with the goal to implement, in close collaboration with academia and industry, efficient and practical solutions for crash simulations, among others.