Argonne Training Program on Extreme-Scale Computing (ATPESC)

By | Educational, Events, HPC, News

Annual Argonne Training Program on Extreme-Scale Computing (ATPESC) 2023:

  • Deadline to submit applications: March 1, 2023 (midnight, Anywhere on Earth)
  • Expected notification of acceptance: May 16, 2023
  • Program start and end date: July 30-August 11, 2023
  • Place: Chicago, IL
  • More information

Submit your application for an opportunity to learn the tools and techniques needed to carry out computational science on the world’s most powerful supercomputers. ATPESC participants will have access to DOE’s leadership-class systems at the ALCF, OLCF, and NERSC.

Program curriculum:

Renowned scientists and leading HPC experts will serve as lecturers and guide the hands-on training sessions. The core curriculum will cover:
Computer architectures
Numerical algorithms and mathematical software
Software productivity and sustainability
Data analysis, visualization, I/O, and methodologies for big data applications
Performance measurement and debugging tools
Machine learning and data science

Eligibility and application:

Doctoral students, postdocs, and computational scientists interested in attending ATPESC can review eligibility and application details on the website.

There are no fees to participate in ATPESC. Domestic airfare, meals, and lodging are also provided.

Los Alamos National Laboratory’s X Computational Physics Workshop and Internship Programs 2023

By | Educational, Events, News, SC2 jobs

Applications for LANL’s X-Computational Physics (XCP) division’s summer 2023 workshop / internship programs are open now. Participants will receive a fellowship stipend, the amount to be determined based on your current academic rank. There are two programs, the computational physics workshop and the parallel computing workshop. Both programs are 10 weeks in duration and require US citizenship. Admissions are rolling, with a closing date of 1/16/2023 for the Computational physics workshop and 1/20/2023 for the parallel computing workshop.

The workshops are geared toward undergraduates and early graduate students, and either would be a great introduction to LANL, and aspects of XCP’s mission (there is significant lecture time built into the schedule of both). Please note that these are effectively XCP Divisions summer internship program if students don’t already have a direct hire mechanism worked out.

Computational Physics Summer Workshop 2023

Date: June 12 – August 18, 2023

Application deadline: January 16, 2023

More information.

 

Parallel Computing Summer Research Internship 2023

Date: June 12 – August 18, 2023

Application deadline: January 20, 2023

More information.

4th Rising Stars in Computational and Data Sciences workshop

By | Educational, Events, News, Research

The Oden Institute for Computational Engineering and Sciences at UT Austin, Sandia National Laboratories (SNL), and Lawrence Livermore National Laboratory (LLNL) are partnering to host the 4th Rising Stars in Computational and Data Sciences, an intensive workshop for women graduate students and postdocs who are interested in pursuing academic and research careers.

Date: April 12-13, 2023

Place: Austin, TX University of Texas Oden Institute for Computational Engineering and Sciences

The Oden Institute is seeking nominations for outstanding candidates in their final year of PhD or within three years of having graduated. We will select approximately 30 women to come for two days of research presentations, poster sessions, and interactive discussions about academic and research careers, with financial support for travel provided.

The nomination consists of sending (1) a letter of nomination and (2) a copy of the nominee’s 2-page resume to Karissa Vail at karissa.vail@austin.utexas.edu. More information can be found at https://risingstars.oden.utexas.edu/

Please consider nominating one of your outstanding current/recent PhD students or postdocs.

Nominations are due:  January 23, 2023.

Tutorial Workshop – Isaac Newton Institute

By | Educational, Events, News

Workshop theme

This week-long workshop will provide an introduction to the core theoretical and applied engineering topics of the DDE programe. Three sessions will be devoted to foundational techniques and methodologies including (i) reinforcement learning and control, (ii) uncertainty quantification and data assimilation, and (iii) model reduction. Another three sessions will be devoted to introductions to current and future challenges for data-driven engineering arising from (i) aeronautics, (ii) chemical and (iii) structural engineering. The tutorials will lay the foundation for the follow-up activities of the DDE programe and the deep-dive study periods in particular. The tutorial workshop is suited in particular for PhD students, postdocs and earlier career scientists from mathematics, statistics, computer science, and computational engineering.

Speakers

  • Sean Meyn (Algorithm design for reinforcement learning and optimization)
  • Sebastian Reich (Uncertainty quantification and data assimilation)
  • Karthik Duraisamy (Model order reduction for complex systems)
  • Luca Magri (Physics aware machine learning in engineering),
  • Antonio del Rio Chanona (Process control and supply chain optimization)
  • Elizabeth Cross (Data-driven structural assessment)

Date

January 16 – 20, 2023

Location

Cambridge, UK or Online (more information in the application section)

More information

Application Deadline: 30 Sep 2022

GRADUATE STUDENT RESEARCH ASSISTANT POSITION IN MECHANICAL ENGINEERING, UNIVERSITY OF MICHIGAN

By | Educational, SC2 jobs

Prof. Jesse Capecelatro in the Department of Mechanical Engineering is currently seeking a GSRA for a DOE-sponsored project related to methane ares.

Job Description

The project will involve performing CFD simulations of methane ames in the presence of turbulent cross-winds and using machine learning to guide improved burner designs. The student must be admitted in College of Engineering or Applied Math and have experience using OpenFOAM (or related CFD) and course work in turbulence and combustion.

 

How to Apply

If you are interested in this position or want to learn more, please email your curriculum vitae to Jesse Capecelatro at jcaps@umich.edu with subject line REM-EDY.

POSTDOC POSITION IN MECHANICAL / NAVAL ARCHITECTURE AND MARINE ENGINEERING, UNIVERSITY OF MICHIGAN

By | Educational, Funding Opportunities, SC2 jobs

Dr. Harish Ganesh (Department of Naval Architecture and Marine Engineering), Prof. Jesse Capecelatro (Department of Mechanical Engineering), and Prof. Steven Ceccio (Department of Mechanical Engineering) are currently seeking a post-doctoral scholar for a one-year position.

Job Description:

This position involves developing numerical methods to leverage time-resolved 3-D experimental measurements using Particle Image Velocimetry (PIV) and Particle Tracking Velocimetry (PTV). Specifically, the project aims at developing a particle based numerical framework to augment experimentally obtained PTV data in the near- and far-wake behind a bluff body. The candidate should have a PhD in Engineering or a related field (e.g., Physics, Mathematics, or Computer Science) with experience in Scientific Computing (proficiency in MPI and Python) and an interest in high Reynold’s number turbulent flows.

How to Apply:

If you are interested in this position, please email your curriculum vitae and at least two references to Jesse Capecelatro at jcaps@umich.edu.

Postdoc Position at Johns Hopkins University

By | Educational, Funding Opportunities, SC2 jobs

Profs. Michael Shields and Lori Graham-Brady are seeking a postdoctoral scholar for a joint collaborative program on uncertainty quantification of mechanical property predictions based on data-driven and/or machine learning models. In the short term, the focus of this project will be on developing UQ tools for Deep Neural Network models that connect image data to material properties. In the longer term this is expected to expand to a spectrum of ML and other data-driven models. Finally, we hope to explore UQ for interconnected models, e.g., in a hierarchical multi-scale construct.

Successful candidates for the position will have a demonstrated track record of scholarly research, and experience in uncertainty quantification, machine learning, and/or computational mechanics. Due to the broad range of expertise requested, we are open to candidates who have more significant expertise in one of these areas and who are interested in developing further knowledge in the other domains. Initial appointment is for one year with the expectation of renewal for a second year pending satisfactory performance. We are committed to recruiting a diverse community of faculty, students, and staff, and to cultivating an inclusive environment that supports, fosters and celebrates all the ways in which the broad differences among us make us better.

Interested candidates are asked to submit their CV and a brief cover letter to Profs. Graham-Brady and Shields. Please email materials to both lori@jhu.edu and michael.shields@jhu.edu.

Professor Karthik Duraisamy named new director of the Ph.D. in Scientific Computing program

By | Educational, Feature
Prof. Karthik Duraisamy infront of screen with turbulence simulation

Professor Karthik Duraisamy (Aerospace Engineering)

Karthik Duraisamy, associate professor of Aerospace Engineering, and an associate director of the Michigan Institute for Computational Discovery & Engineering, has been named director of the joint Ph.D. in Scientific Computing program effective on January 1, 2022. Professor Duraisamy’s research involves the development of theory and algorithms for computational modeling of complex physical systems. He was the principal investigador of ConFlux, an NSF Major Research Instrumentation project that led to the development of a first of its kind computing instrument specifically designed to enable High Performance Computing (HPC) clusters to communicate seamlessly and at interactive speeds with data-intensive operations. Currently he directs the Air Force Center of Excellence on Rocket Combustion modeling. He is invested in educating future researchers with a strong computational background capable of using the power of computing for problem solving. He worked with the group that launched the course Methods and Practice in Scientific Computing, and developed and teaches a course on data-driven analysis and modeling of complex systems. These two courses give students a solid foundation, enabling them to use HPC in their research. 

Portrait of Ken Powell

Professor Ken Powell (Aerospace Engineering)

Professor Duraisamy replaces Ken Powell, Arthur F. Thurnau Professor of Aerospace Engineering, who stepped down from the role after 18 years of service. As a young assistant professor, Professor Powell was an instrumental member of the original team that conceived and launched the program back in 1989. The field of computational fluid dynamics, where his research interests lie, has always included an active community of HPC users and developers, thus he was always actively involved in the program through research, teaching and student advising. In 2004 he succeeded Professor William Martin as director of the program. During his time as director, he met and advised every single one of the over 350 students that enrolled in the program. Through this time he became an expert on scientific computing courses across the university, and witnessed first hand the explosion in computational and data science usage, reflected in the research scope of the students enrolling in the program.

Professor Duraisamy has big shoes to fill, but he is being assisted by the MICDE Management and Education Committee. The program’s mission, to train U-M students in scientific computing and to support the growing computational and data science community at the University of Michigan, will itself continue to expand.

The University of Michigan Ph.D. in Scientific Computing timeline. Read more.

 

XSEDE HPC Workshop: BIG DATA and Machine Learning

By | Educational, Events, HPC

XSEDE and the Pittsburgh Supercomputing Center are offering a two day Big Data and Machine Learning virtual workshop that will focus on topics including big data analytics and machine learning with Spark, and deep learning using Tensorflow.

When: Wed., April 6 @ 11:00 a.m. – 5:00 p.m. E.S.T. & Fri., April 8 @11:00 a.m. – 5:30 p.m. E.S.T.

Registration closes on April 4, 2022. Space is limited.

Tentative Agenda

Wednesday, April 6
All times given are Eastern
11:00 Welcome
11:25 A Brief History of  Big Data
12:20 Intro to Spark
1:00    Lunch break
2:00    More Spark and Exercises
3:00    Intro to Machine Learning
5:00    Adjourn

Friday, April 8
All times given are Eastern
11:00 Machine Learning: Recommender System with Spark
1:00    Lunch break
2:00    Deep Learning with Tensorflow
5:00    Tying it All Together
5:30    Adjourn

Applications to the 2022 Annual Argonne Training Program on Extreme-Scale Computing are due March 1

By | Educational, HPC

Argonne Training Program on Extreme-Scale ComputingThe annual Argonne Training Program on Extreme-Scale Computing (ATPESC), for doctoral students, postdocs, and computational scientists, is set to take place July 31-August 12, 2022. This year’s program will mark the 10th anniversary of ATPESC.

Submit your application for an opportunity to learn the tools and techniques needed to carry out computational science on the world’s most powerful supercomputers. ATPESC participants will have access to DOE’s leadership-class systems at the ALCF, OLCF, and NERSC.

Call for 2022 Applications extended to March 7

Learn more and apply here

PROGRAM CURRICULUM
Renowned scientists and leading HPC experts will serve as lecturers and guide the hands-on sessions. The core curriculum will cover:

  • Hardware architectures
  • Programming models and languages
  • Approaches for performance portability
  • Numerical algorithms and mathematical software
  • Performance measurement and debugging tools
  • Data analysis, visualization, and methodologies for big data applications
  • Approaches to building community codes for HPC systems
  • Machine learning and data science

ELIGIBILITY AND APPLICATION
Doctoral students, postdocs, and computational scientists interested in attending ATPESC can review eligibility and application details on the website.

There are no fees to participate in ATPESC. Domestic airfare, meals, and lodging are also provided.
Application deadline: March 1, 2022.

ATPESC is funded by the Exascale Computing Project, a collaborative effort of the DOE Office of Science’s Advanced Scientific Computing Research Program and the National Nuclear Security Administration.