CC* Data Storage: Cost-effective Attached Storage for High throughput computing using Homo- geneous IT (CASH HIT) supporting Penn State Science, the Open Science Grid and LIGO

Scientific computing needs continue to grow at an exponential rate driven by an exponential growth in data that often outpaces available resources. High performance, cost effective, data storage is a critical component of scientific workflows that produce and analyze large datasets. Penn State is providing cost-effective data storage to researchers across many scientific domains including physics, astronomy, biology and materials science. This work supports science at various scales from local Penn State researchers to researchers across the country. Much of the storage is available to external communities through the Open Science Grid, which allows US researchers to gain free access to computing resources. The project team is exploring several important scientific topics with storage including searching for ripples in space caused by black holes, searching for planets outside of our solar system that could support life, and studying gene regulation. Project goals include developing relationships with national cyberinfrastructure efforts such as the Partnership to Advance Throughput computing; building a workforce that is able to leverage cost-effective, open source storage; and enabling research that otherwise would not be possible with existing storage resources and budgetary constraints. The project aims to provide 7.8 petabytes (PB) of storage. 2.9 PB of the storage (37%) will be available to external communities while the remaining 4.9 PB will be used internally. Among the externally allocated storage, 2 PB will be used for Open Science Data Federation (OSDF) applications and 900 TB will support a new regional initiative to provide storage for underserved institutions in Pennsylvania. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.