Mount Sinai has committed over $50 million to its scientific computational and data cyber infrastructure. This is a substantial commitment by Mount Sinai that demonstrates, in a very tangible manner, the understanding that a well-designed and managed infrastructure empowers scientists and researchers to be more productive and effective. Furthermore, this substantial investment is a holistic one that includes a professional, operational and experienced scientific computing staff along with the hardware, software and new computational and data facilities optimized for our scientific workload.
Mount Sinai's current computational resource, named Minerva after the Roman goddess of medicine and science, occupies a portion of the 30,000 square feet available in our computing facility. Minerva incorporates 7,680 AMD 2.3 GHz Interlagos cores in 120 4-socket compute nodes each with 256 GB of memory, for a total system memory of 30 terabytes providing over 100 teraflops of peak computational power to our scientists. A significant upgrade is planned for Q4 2013 that will more than double the size of Minerva. Special purpose nodes are available with NVIDIA GPUs, 1 TB of shared system memory, MATLAB Distributed Computing Server, and web and database services. The storage subsystem contains 1.5 petabytes of high performance parallel file system storage and will expand to over 5 petabytes before the end of 2013. An archival storage system encrypts and saves copies of data on tape to two geographically disparate locations.
The software and programming environments are the best of breed, and include community standards such as Linux and MPI. Over 600 scientific software packages are available and regularly updated on Minerva. The workload on the Minerva is controlled by resource managers and schedulers that optimize utilization for maximum job throughput or time to solution based on the priorities of our scientific community. Minerva achieves well over 99% uptime, using scalable and reproducible configuration management techniques. Rigorous and industry-standard multi-factor authentication and other security policies ensure the integrity of the researcher's data and Mount Sinai's compute infrastructure.
Mount Sinai also employs several domain-specific PhD-level computational scientists with interdisciplinary expertise to assist researchers to make efficient and effective use of computing resources. These scientists lead basic and advanced training classes and accelerate the scientific discovery process by assisting with code development, optimization and troubleshooting. They also conduct annual surveys and actively solicit and respond to feedback.
Additionally, Mount Sinai's robust computing and data cyber infrastructure has been designed for the rapid and accurate ingest of the sequencer output, and high performance post-processing and analysis by the computational and storage infrastructure. The cyber infrastructure resources have been tailored specifically to handle the computational and data workflow from the sequencer, including a high bandwidth network to the high performance computing and data facilities.
Mount Sinai collaborates and partners with other facilities and vendors to continuously improve its state-of-the-art cyber infrastructure and services. The scientific computing staff track the communities' best practices and procedures to ensure that Mount Sinai's computing and data services are efficient and effective for its researchers.
Scientific Computing [PDF]