Browsing by Author "Michael, Scott"

Now showing 1 - 14 of 14

2016 UITS Annual Survey Overview
(2016-11-03) Wernert, Julie; Michael, Scott; Gniady, Tassie
Acceptance Test Results for the Jetstream2 Production Environment
(2022-04-19) Hancock, David Y.; Turner, George; Weakley, Le Mai; Teige, Scott; Lowe, J. Michael; Bird, Stephen; Michael, Scott; Martin, Chris; Fischer, Jeremy; Snapp-Childs, Winona; Pierce, Marlon; Marru, Suresh; Wannipurage, Dimuthu
Advanced Topics in R
(Indiana University Workshop in Methods, 2014-12-05) Davis, Jefferson; Zhang, Hui; Michael, Scott
In this follow up workshop to “Introduction to R” researchers will have the opportunity for a deeper dive into R. Available on all IU's supercomputers, R is a flexible open source statistical programming language that can work with large and complex data sets. This workshop will address several advanced topics in R as well as giving participants the opportunity to use R on IU’s supercomputers. The topics to be covered include: R scripting on IU supercomputers; debugging, profiling, and performance analysis of R code; parallel programming in R, including the Rmpi and snowfall packages; and advanced plotting in R. Participants will have access to the supercomputer Big Red II during the session and will be able to see hands-on examples of running R code and submitting batch jobs in R.
AI and computing: Local tools @IU, some global challenges
(2019-09-11) Michael, Scott; Stewart, Craig A.
Creating Computational Interfaces
(2018-03-20) Michael, Scott; Gniady, Tassie
This workshop will focus on Jupyter/Zeppelin notebooks and Shiny R interfaces, and how to develop and where to run them.
Indiana University and Purdue University Remote Desktop User Assessment Study 2019
(2019-02-15) Thota, Abhinav; Weakley, Le Mai; Fulton, Ben; Dennis, H.E. Cicada Brokaw; Huber, Laura; Michael, Scott; Snapp-Childs, Winona; Wernert, Julie; Younts, Alexander; Harrell, Stephen Lien; Dietz, Daniel T.; Phillips, Christopher; Zhu, Xiao
Migration of Gas Giant Planets in Gravitationally Unstable Disks
(2011-05-25) Michael, Scott; Durisen, Richard H.; Boley, Aaron C.
Characterization of migration in gravitationally unstable disks is necessary to understand the fate of protoplanets formed by disk instability. As part of a larger study, we are using a 3D radiative hydrodynamics code to investigate how an embedded gas giant planet interacts with a gas disk that undergoes gravitational instabilities (GIs). This Letter presents preliminary results from simulations with a Jupiter-mass planet placed in orbit at 25 AU within a 0.14 M_sun disk. The disk spans 5 to 40 AU around a 1 M_sun star and is initially close to marginal stability. In one simulation, the planet is inserted prior to the eruption of GIs; in another, it is inserted only after the disk has settled into a quasi-steady GI-active state, where heating by GIs roughly balances radiative cooling. When the planet is present from the beginning, its own wake stimulates growth of a particular global mode with which it strongly interacts, and the planet plunges inward six AU in about 1000 years. In both cases with embedded planets, there are times when the planet's radial motion is slow and varies in direction. At other times, when the planet appears to be interacting with strong spiral modes, migration both inward and outward can be relatively rapid, covering several AUs over hundreds of years. Migration in both cases appears to stall near the inner Lindblad resonance of a dominant low-order mode. Planet orbit eccentricities fluctuate rapidly between about 0.02 to 0.1 throughout the GI-active phases of the simulations.
Performance Characteristics of Virtualized GPUs for Deep Learning
(2019-10) Michael, Scott; Teige, Scott; Li, Junjie; Lowe, John Michael; Turner, George; Henschel, Robert
As deep learning techniques and algorithms become more and more common in scientific workflows, HPC centers are grappling with how best to provide GPU resources and support deep learning workloads. One novel method of deployment is to virtualize GPU resources allowing for multiple VM instances to have logically distinct virtual GPUs (vGPUs) on a shared physical GPU. However, there are many operational and performance implications to consider before deploying a vGPU service in an HPC center. In this paper, we investigate the performance characteristics of vGPUs for both traditional HPC workloads and for deep learning training and inference workloads. Using NVIDIA’s vDWS virtualization software, we perform a series of HPC and deep learning benchmarks on both non-virtualized (bare metal) and vGPUs of various sizes and configurations. We report on several of the challenges we discovered in deploying and operating a variety of virtualized instance sizes and configurations. We find that the overhead of virtualization on HPC workloads is generally < 10%, and can vary considerably for deep learning, depending on the task.
Planet Migration Induced by Gravitational Instabilities
(Indiana University, 2011-02-18) Michael, Scott
The study of gravitational instabilities (GIs) and their effects in protoplanetary disks has been an area of active research for over a decade. Although some studies have indicated that GIs cannot form gas giant planets directly, it is clear that they can have a significant effect on a protoplanetary disk. In this dissertation I present several areas where GIs may play a key role in the evolution of a protoplanetary disk. These studies are carried out using three dimensional numerical simulations. I have carried out and analyzed nearly twenty simulations with varying initial conditions, resolutions, and physical effects. Although all indications from these simulations are that GIs cannot form gas giant planets directly at radii smaller than 40 AU, they have shown that GIs can have a dramatic effect on protoplanetary disk structure and planets embedded in a protoplanetary disk. I present several key results including: the effects of a varied initial surface density profile azimuthal resolution, the amplitude of the initial random perturbation, and the adiabatic index used on the onset, strength and general evolution of GIs in protoplanetary disks. Additionally, I present results on studies of the interaction of the instabilities with the central star when it is allowed to move freely in response to the action of the GIs. Finally, I present several results regarding the interaction of embedded massive planets and GI active disks. I find that the presence of massive planets can have a dramatic effect on the evolution of GIs in an active disk, and the GIs can also dramatically effect them migration of the embedded planet. In fact, the action of the GIs may planets to migrate outward, contrary to the standard theory of the planet migration in laminar disks.
Security standards compliance and ease of use of high performance computing systems in clinical research
(2021-07) Link, Matthew; Shankar, Anurag; Hancock, David Y.; Henschel, Robert; Michael, Scott; Stewart, Craig A.
Precision health research and personalized health therapies involve analysis of protected health information. In 2007, Indiana University established the ability to analyze protected health information (HIPAA alignment) as the minimal and default security level for its research high performance computing (HPC) systems and research storage systems. This resulted in a drastic increase in the use of IU HPC systems by clinical researchers. Security levels were later upgraded to FISMA Low as a default. We recommend that, within the US, FISMA (Federal Information Security Modernization Act) Low compliance be the default minimal level of security for large-scale HPC systems. This would facilitate precision medicine research and enable higher education HPC resources to be used in response to future civil health emergencies.
Services and support for IU School of Medicine and Clinical Affairs Schools by the UITS/PTI Advanced Biomedical Information Technology Core and Research Technologies Division in FY 2013 - Extended Version
(Indiana University: UITS/PTI ABITC and RT Division, 2013-12-19) Stewart, Craig A.; Barnett, William K.; Link, Matthew R.; Shankar, Ganesh; Miller, Therese; Michael, Scott; Henschel, Robert; Boyles, Michael J.; Wernert, Eric; Quick, Robert
The report presents information on services delivered in FY 2013 by ABITC and RT to the IU School of Medicine and the other Clinical Affairs schools that include the Schools of Nursing, Dentistry, Health and Rehabilitation Sciences, and Optometry; the Fairbanks School of Public Health at IUPUI; the School of Public Health at IU Bloomington; and the School of Social Work.
Technical Report: Report on Lustre use across an experimental 100Gb network spanning 2,175 mi
(2012) Henschel, Robert; Simms, Stephen C.; Hancock, David Y.; Michael, Scott; Johnson, Tom; Heald, Nathan; William, Thomas; Berry, Donald; Allen, Matt; Knepper, Richard; Davy, Matthew; Link, Matthew R.; Stewart, Craig A.
Indiana University demonstrated use of the Lustre high performance parallel file system over an experimental 100 Gbps network made available as part of the SCinet Research Sandbox at the Supercomputing 2011 conference. The network distance was 2,175 miles which yielded a latency of 50.5 ms. This report outlines the findings from low level iperf network tests and file system tests with the IOR benchmark to a suite of real-world applications reading and writing to the file system.
Usage of Indiana University computation and data cyberinfrastructure in FY 2011/2012 and assessment of future needs
(2013-06-07) Link, Matthew R.; Hancock, David Y.; Seiffert, Kurt; Simms, Stephen; Michael, Scott; Stewart, Craig A.
This report details the past and current cyberinfrastructure resources that have been deployed by the Research Technologies (RT) division of University Information Technologies Services to support research and scholarly activities at IU. This report also presents data and detailed analysis of system usage and services supported by RT for the FY 2011/2012 period, projects future usage trends based on these data, and provides several recommendations for the most effective ways to meet the growing need for high performance computing resources in research and scholarly endeavors.
Who Cares About Science Gateways? A Large-Scale Survey of Community Use and Needs (Dataset)
(2015-09-17) Lawrence, Katherine A.; Zentner, Michael; Wilkins-Diehr, Nancy; Wernert, Julie A.; Pierce, Marlon; Marru, Suresh; Michael, Scott
With the rise of science gateway use in recent years, we anticipate there are additional opportunities for growth, but the field is currently fragmented. This data set represents the results our efforts to measure the extent and characteristics of the gateway community through a large-scale survey. Our goal was to understand what type of support services might be provided to the gateway community.