gatewayconf2017.pdf (8.02 MB)

Unidata Science Gateway on the XSEDE Jetstream Cloud

Download (8.02 MB)
posted on 06.10.2017 by Julien Chastang, Mohan Ramamurthy
With the goal of better serving our community and in fulfillment of objectives articulated in "Unidata 2018: Transforming Geoscience through Innovative Data Services," Unidata is investigating how its technologies can best make use of cloud computing. The observation that science students and professionals are spending too much time distracted by software that is difficult to access, install, and use, motivates Unidata’s investigation. In addition, cloud computing can tackle a class of problems that cannot be approached by traditional, local computing methods because of its ability to scale and its capacity to store large quantities of data. Cloud computing accelerates scientific workflows, discoveries, and collaborations by reducing research and data friction. We aim to improve “time to science” with the NSF-funded XSEDE Jetstream cloud. We describe a Unidata science gateway on Jetstream. With the aid of open-source cloud computing projects such as OpenStack and Docker on Linux VMs, we deploy a variety of scientific computing resources on Jetstream for our scientific community. These systems can be leveraged with data-proximate Jupyter notebooks, and remote visualization clients such as the Unidata Integrated Data Viewer (IDV) and AWIPS CAVE. This gateway will enable students and scientists to spend less time managing their software and more time doing science.