Hybrid HPC and Big Data Cluster


For meeting 21st century’s big data challenge and to advance data science research, ACER now provides UIC researchers access to hybrid HPC/Big-Data resource called SABER, A Shared Analytics and Big Data Enterprise Resource. Funded by a National Science Foundation (NSF) grant, SABER is UIC’s first fee-for-service grant where researchers can either buy service time or buy dedicated nodes for exclusive usage. With over one petabyte of research storage, SABER will allow researchers to run computation and analysis on massive datasets from heterogeneous sources. Hardware configuration (See SABER specs) and technical documentation along with FAQs can be found on the right side of this page.

SABER is now operational and available campus-wide on a fee-for-service basis. To facilitate the transition from desktop computing to cluster computing, our team will work with each research group to install/build its essential software and provide 500 core-hours at no charge with 25 GB of storage. Additional storage can be purchased incrementally. Detailed information about pricing and ordering process is located to the right of this page.

Big Data workloads can be processed on SABER using its Hadoop On Demand (HOD) service. Using HOD, MapReduce jobs can operate on dynamically-provisioned Hadoop clusters through existing HPC job scheduler services. This eliminates the need for researchers to acquire and configure their own dedicated resources for Big Data analysis.

Please contact us at to learn more about SABER project and its availability to campus community.