In addition to PTI, a seasoned team of organizations and experts will implement Jetstream.
TACC designs and deploys the world's most powerful advanced computing technologies and innovative software solutions to enable researchers to answer complex questions like these and many more. Every day, researchers rely on TACC computing experts and resources to help them gain insights and make discoveries that change the world.
TACC’s scientific computing environment includes a comprehensive cyberinfrastructure ecosystem of leading-edge resources in high performance computing (HPC), visualization, data analysis, storage, archive, data-driven computing, connectivity, tools, APIs, algorithms, consulting, research cloud, and software. In addition, its skilled experts work with thousands of researchers on more than 3,000 projects each year. With Jetstream, TACC adds a powerful production science cloud for science and engineering to its extensive portfolio of capabilities that is expected to expand even further TACC’s impact in research computing.
"We've identified a serious need among current and potential XSEDE users for the flexible software environments, enhanced research reproducibility, and ease of use that can be provided by cloud computing," said Matt Vaughn, a co-principal investigator of iPlant Collaborative and director of life sciences computing at TACC. "Our experience with the success of iPlant's Atmosphere platform, which features resources hosted both at University of Arizona and at TACC, is that users will surprise us by how quickly they take to using Jetstream and how much world-changing research they'll get done. We're excited to get started building it."
The University of Chicago Computation Institute will be responsible for integrating Jetstream user authentication and data movement with the NSF-funded XSEDE (Extreme Science and Engineering Discovery Environment) program. Dr. Ian Foster, one of the most important computer scientists in the US, will lead the Computation Institute in integration of its renowned suite of Globus data movement and management tools, as well as Globus-based secure user authentication.
Cloud computing has fulfilled the need for developing highly customized, scalable analysis platforms that are essential infrastructure components for researchers today. Wide adoption of these cloud-based capabilities at iPlant is driven in part by use of Atmosphere, a multi-cloud orchestration platform developed by the iPlant Collaborative; it provides an easy-to-use web interface for researchers to manage their cloud-based assets and analyses with ease.
iPlant Atmosphere will manage the cloud resources and provisioning of the production environments at Indiana University and the Texas Advanced Computing Center, providing the familiar web interface for the estimated 45,000 Jetstream early adopters. The University of Arizona will host test and development facilities for Jetstream. The iPlant Collaborative is an NSF-funded project providing cyberinfrastructure for life sciences research. Since January 2011, iPlant Atmosphere has managed production cloud services for iPlant. Atmosphere provides a software platform to ease the use and management of cloud resources by scientists, greatly increasing the adoption of cloud computing in research applications.
"We have seen overwhelming demand for our cloud platform from researchers and educators alike," said Nirav Merchant, co-principal investigator of iPlant Collaborative, director of bio computing at Arizona Research Laboratories and member of the BIO5 Institute at the University of Arizona. "With Jetstream we now have an avenue for our users needing more capacity. Jetstream provides the opportunity to take best practices from our experience customizing cloud infrastructure for life sciences to a broader national community."
The Cloud and Big Data Laboratory of UTSA, led by Paul Rad, will be Jetstream’s key partner in testing and implementing new features in OpenStack as they become available. Unlike a traditional supercomputer, which is accepted once and then stays largely static for several years other than modest software upgrades, cloud system software tends to change rapidly. The Jetstream team will depend on the UTSA Cloud and Big Data Laboratory for support and maintenance of the OpenStack software that powers Jetstream.
"Cloud is at the earliest stage of adoption," said Paul Rad, director of applied research at the University of Texas, San Antonio. "It's a phenomenon that is going to change the shape not only of industry but of our lives. We at the UTSA Open Cloud Lab, supported by Rackspace who co-founded OpenStack with NASA, are very excited to be working closely with Indiana University and UT TACC to build the ‘super cloud’ for science and engineering."
Galaxy is a software framework and web-based science gateway for data-intensive biology. Since 2005 it has made large-scale data analysis, particularly the analysis of genomic data, accessible to tens of thousands of researchers. A major goal of Galaxy is to address the increasingly recognized reproducibility crisis in computational science by providing automatic provenance for all analyses. However, increasing dataset size and complexity of analyses has meant that Galaxy users face throughput limitations, directly affecting the rate of progress in genomic research.
Jetstream will provide a massive increase in the amount and flexibility of computational resources available to the Galaxy gateway, enabling existing users to perform analysis faster, to perform new kinds of analysis, as well as to support many more users. Galaxy is developed by co-PI James Taylor at Johns Hopkins University in collaboration with Anton Nekrutenko at Penn State University and a vibrant open-source community.
Cornell University is Jetstream’s training partner responsible for providing education to the broader community.
"As an early cloud adopter and provider with Red Cloud, Cornell University Center for Advanced Computing (CAC) staff will leverage their cloud computing expertise to develop and deploy on-demand Jetstream training," said Susan Mehringer, assistant director, CAC. "The first Cornell Virtual Workshop module will introduce users to Jetstream capabilities and teach them how to optimally use, create, and archive cloud services and virtual machines (VMs)," she explained. Subsequent modules will focus on using remote desktops to access Jetstream, XSEDE, and systems in the XD program; deploying biology and earth science applications on Jetstream; and, publishing, archiving, and curating VM images to ensure scientific reproducibility.
Dr. Jessee Walker of University of Arkansas at Pine Bluff will develop and implement tools for cybersecurity education using Jetstream. He will also advise Jetstream in promoting adoption of Jetstream at minority-serving universities and colleges throughout the US.
The National Snow and Ice Data Center is the definitive US repository of information about the world’s glaciers. NSIDC will work with Jetstream to make these data and tools to analyze these data available through Jetstream.
The Odum Institute at the University of North Carolina will advise the Jetstream team in deploying tools related to the social sciences.
Dr. Gwen Jacobs of the University of Hawaii will chair the Jetstream Science Advisory Board and be a key leader in setting Jetstream science and engineering support strategies. As a noted cyberinfrastructure leader in an Experimental Program to Stimulate Competitive Research (EPSCoR) state, Dr. Jacobs will be particularly influential in steering Jetstream strategies to aid researchers and educators in states that receive relatively little NSF funding. In addition, the University of Hawaii and Jetstream will collaborate to support oceanography and genomics research by the University of Hawaii.
Dell will provide the hardware for Jetstream, which will consist of two homogenous clusters at Indiana University and TACC with a test environment at the University of Arizona.
"Dell is committed to enabling innovation and discovery," said Jimmy Pike, vice president, senior fellow and chief architect, Dell. "Our HPC solutions are deployed across the globe as the computational foundation for academic and governmental research critical to scientific advancement. This partnership demonstrates the value of Dell’s innovative solutions in building campus-based clouds for academic research, and we are excited about the insight Jetstream and those managing the cloud will uncover."