Search for: All records

High-performance computing (HPC) is widely used in higher education for modeling, simulation, and AI applications. A critical piece of infrastructure with which to secure funding, attract and retain faculty, and teach students, supercomputers come with high capital and operating costs that must be considered against other competing priorities. This study applies the concepts of the production function model from economics with two thrusts: (1) to evaluate if previous research on building a model for quantifying the value of investment in research computing is generalizable to a wider set of universities, and (2) to define a model with which to capacity plan HPC investment, based on institutional production—inverting the production function. We show that the production function model does appear to generalize, showing positive institutional returns from the investment in computing resources and staff. We do, however, find that the relative relationships between model inputs and outputs vary across institutions, which can often be attributed to understandable institution-specific factors. 

Over the past decade, the convergence of Cloud and High-Performance Computing (HPC) has undergone significant movement. We explore the evolution, motivations, and practicalities of establishing on-premise research cloud infrastructure and the complementary nature with HPC and commercial resources; under the belief that research clouds serve a unique role within research and education as a convergence accelerator. This role is highlighted through exploring the design tradeoffs in architecting research clouds versus HPC resources, focusing on the balance between utility, availability, and hardware utilization. The discussion provides insights from experiences with the National Science Foundation-supported Jetstream and Jetstream2 systems, showcasing convergence technologies and challenges. A variety of real-world use cases are provided that show the interplay between these computing paradigms; exploring use in research and education for interactive and iterative development, as an on-ramp to large-scale resources, as a powerful tool for education and workforce development, and for domain specific science gateways. 

Jetstream2 will be a category I production cloud resource that is part of the National Science Foundation’s Innovative HPC Program. The project’s aim is to accelerate science and engineering by providing “on-demand” programmable infrastructure built around a core system at Indiana University and four regional sites. Jetstream2 is an evolution of the Jetstream platform, which functions primarily as an Infrastructure-as-a-Service cloud. The lessons learned in cloud architecture, distributed storage, and container orchestration have inspired changes in both hardware and software for Jetstream2. These lessons have wide implications as institutions converge HPC and cloud technology while building on prior work when deploying their own cloud environments. Jetstream2’s next-generation hardware, robust open-source software, and enhanced virtualization will provide a significant platform to further cloud adoption within the US research and education communities. 

Abstract Neuroscience is advancing standardization and tool development to support rigor and transparency. Consequently, data pipeline complexity has increased, hindering FAIR (findable, accessible, interoperable and reusable) access. brainlife.io was developed to democratize neuroimaging research. The platform provides data standardization, management, visualization and processing and automatically tracks the provenance history of thousands of data objects. Here, brainlife.io is described and evaluated for validity, reliability, reproducibility, replicability and scientific utility using four data modalities and 3,200 participants.