Speaker: Daniel Howard, NCAR/CISL

Abstract

Research Software Engineers, or RSEs, are vital to the modern research enterprise. Nearly all instrumentation and analysis workflows are dependent on software. Within HPC, modeling software itself is considered just as much the scientific instrument as the cyberinfrastructure hardware it runs on. Given that NSF NCAR plays an integral role in advancing science and technology for the public good, it is important to recognize the value people who serve in RSE roles bring towards enabling the long term sustainability, innovation, and usability of research products, such as through CISL's SEAL initiative, Software Engineering Across Labs.

This seminar aims to highlight the value RSEs bring to the scientific community and how you can get involved. Specifically, the ongoing work of the US-RSE organization, now 2,500+ strong, will be discussed. Being free to join, opportunities to engage will be shared including various volunteer opportunities and conferences. Members may also join working groups, focused on areas including code review, user experience, RSE empowerment in national labs, and others.

Given the software complexity of NSF NCAR's large modeling programs, there is significant interest towards how best to retain and cultivate essential RSE talent who build and innovate software solutions in Earth system science. Organizations must evolve their research enterprise to promote and reward these valuable contributions RSEs bring. The growing RSE community both in the US and globally can certainly support this effort.
 

Biography

Daniel Howard is an HPC Consultant in CISL and joined the Consulting Services Group in 2021. Previously, Daniel worked on GPU development efforts for the NRL's WaveWatch3 program and studied monsoon dynamics at Notre Dame. With a background in applied mathematics and computational science, Daniel is passionate about furthering Earth system science through best practices in research software engineering as well as the use of compute efficient mathematical toolsets deployed on modern computing hardware, which recently has been GPU processors.

Presentation Slides

Seminar Recording

Speaker: Yue (Michael) Ying, Nansen Environmental and Remote Sensing Center (NERSC)

Abstract

The Next-generation Ensemble Data Assimilation System (NEDAS) provides a light-weight Python solution for implementing ensemble data assimilation methods for geophysical models. Thanks to its modular and scalable design, a wide range of ensemble assimilation algorithms become feasible for large-dimensional models. NEDAS provides a collection of state-of-the-art algorithms from existing research software using two main strategies: (1) serial assimilation where observations are used one at a time to update the model state iteratively, (2) batch assimilation where observations are in a local analysis for each model state. One can test new algorithmic ideas in NEDAS and compare them with existing methods early-on before committing resources to full implementation in a real operational setting. In this talk, I’ll describe the architectural design of NEDAS, highlight some new algorithmic ideas, and show some computational benchmarks as a guidance for picking the right algorithm in different application scenarios.

Biography

Yue Ying, aka Michael, is currently working as a DA researcher at the Nansen Environmental and Remote Sensing Center in Norway. He had his PhD in meteorology in PennState studying the predictability of multiscale tropical weather systems, then invented a multiscale alignment” approach to tackle the nonlinearity in DA problems. Now, he continues to develop new DA methods and apply them to the modeling of the Arctic systems.

Presentation Slides

Seminar Recording

Speaker: Brian D. Haymore, University of Utah Center for High Performance Computing

Abstract

A look at how CHPC has been using Student Part-Time employees for the last 2 decades and how things have improved and changed with the development of the campus SUDO program as it came online in 2015.  Areas of focus will include:

• The structure, goals and services the SUDO program provides to the students as well as the departments using the program.
• How CHPC uses the SUDO program to streamline our hiring process.
• How we manage/mentor students and help encourage good student/employee balance.
• What we do to prepare and train mentors to be effective with students.
• What roles, tasks and projects CHPC utilizes students with.
• How the program fits into our workforce development goals as a department.
• What we see as the metrics of success with both the program and our student workers.

Biography

Brian has been with CHPC at the University of Utah for 26 years with a few years prior as a student employee doing sysadmin work on various SGI, DEC, IBM, and Sun systems. He first got involved with Linux in 1995 and it has been central to everything he's done since. Brian's current role is both the development and operations of the University of Utah's systems as well as managing the rest of the HPC admin team. He is heavily involved in CaRCC with the systems-facing track as well as the student workforce development interest group. ACM SIGHPC SYSPROS and the RMACC community are also areas he participates in regularly. For about the last 20 years he's had student workers involved in supporting their systems alongside full-time staff. The student workers are a integral part of their team.

Presentation Slides

Seminar Recording