High Performance Computing System Acquisition: Jetstream - A Self-Provisioned, Scalable Science and Engineering Cloud Environment

Indiana University, Bloomington IN

Investigators

Abstract

High Performance Computing System Acquisition: Jetstream - a self-provisioned, scalable science and engineering cloud environment Jetstream will be a new type of computational research resource open for the national (nonclassified) research community - a data analysis and computational resource that US scientists and engineers will use interactively to conduct their research anytime, anywhere. Jetstream will complement current NSF-funded computational resources and bring a cloud-based system to the NSF computational resources incorporating the best elements of commercial cloud computing resources with some of the best software in existence for solving important scientific problems. This system will enable many US researchers and engineers to make new discoveries that are important to understanding the world around us and will help researchers make new discoveries that improve the quality of life of American citizens. In terms of technical details, Jetstream will be a configurable large-scale computing resource that leverages both on-demand and persistent virtual machine technology to support a much wider array of software environments and services than current NSF resources can accommodate. As a fully configurable "cloud" resource, Jetstream bridges the obvious major gap in the current ecosystem, which has machines targeted at large-scale High-Performance Computing, high memory, large data, high-throughput, and visualization resources. As the open cloud for science, Jetstream will: *Provide "self-serve" academic cloud services, enabling researchers or students to select a VM image from a published library, or alternatively to create or customize their own virtual environment for discipline- or task-specific personalized research computing. *Host persistent VMs to provide services beyond the command line interface for science gateways and other science services. For example, Jetstream will become a primary host of the popular Galaxy scientific workbench and its main datasets, bringing many Galaxy users to the NSF ecosystem from day one. *Enable new modes of sharing computations, data, and reproducibility. *Expand access to the NSF XSEDE ecosystem by making virtual desktop services accessible from institutions with limited resources