Grande-CARES - Education

To become innovation leaders, the next generation of scientists and engineers must receive an education focused on fundamental theories, knowledge, applications, and experiences, particularly in multiphysics problems. This study aims to develop modular course content for non-CS majors to bridge the knowledge gap and develop an integrated curriculum. The development of module(s) will follow the strategic teaching approach proposed by the Teaching and Learning Laboratory (TLL) of MIT [77]. This approach suggests identifying variables, constraints, informed decision making, assessment approach to be followed, and refinements needed for improvements (Fig. 10, adopted from TLL).

Proposed education tasks

1. Interdisciplinary Education / Curriculum

Core course 1 (New): Computational Methods for Multiphysics Problems (3 credit hours) – This new course will provide the knowledge to address multiphysics problems. This course will be designed based on Albany multiphysics software from Sandia. Modules at UG level.
Core course 2 (New): Advanced Computing (3 credit hours) – This new course will explain HPC, GPU, and MPI-X programming models for exascale computing. This course will be designed using the contents of Trilinos software suites from Sandia.
Core course 3 (New): Data-driven analytics and engineering (3 credit hours) – This new course will provide an in-depth understanding of big data, artificial intelligence, machine learning, and UQ used for engineering applications. The course will be designed in a modular way using Dakota and Tensorflow opensource packages. Modules will be created for students at the UG level.
Prescribed courses (6 credit hrs - 2 of the following): High-temperature reactivity, Advanced Mathematics, CFD, FEA, Advanced fluid dynamics, Advanced Structure Mechanics, Numerical Methods, Scientific Computing or as approved by the thesis/dissertation advisors
Other required credit hours will come from electives or as per the student's project requirement

Internship experience: All Rio Grande CARES students will work with scientists at SNL for at least one semester (required for Ph.D. & MS RA) and one internship (required for UG RA)
Interdisciplinary project: The consortium students will work on at least one interdisciplinary research project (Optional for UG RA)
Advising and committee membership: The students will be co-advised by faculty members from two departments and at least one member from Sandia (Ph.D. & MS RA).

2. Interdisciplinary Student Research

All faculty members will build on their experience in leading their respective research groups. For example, the consortium expectations for all students include:

● Research leadership: All graduate students at the dissertation stage will lead at least one interdisciplinary research group meeting.

● Written communication and accountability: Students will maintain an electronic notebook submitted for review bi-weekly. The additional purpose will be to fine-tune/modify instructions and course modules for better understanding and minimizing struggles often students face with self-paced learning.

● Oral and written communication (advanced): All CARES trainees will present at least one peer-reviewed conference. All graduate students must submit at least one paper in a peer-reviewed journal.

● Advanced computing workshops: All students must participate in the following workshops - XSEDE MPI workshop (2 days), XSEDE OpenMP workshop (1 day), GPU programming using OpenACC (1 day), XSEDE Bigdata workshop (2 days), XSEDE Summer boot camp (4 days), Trilinos User Group meeting (TUG), and Albany User Group meeting (AUG). In addition, students should also participate in the hands-on training sessions of the instructors.

● Seminar: All students must participate in all seminar talks hosted by the consortium. The team will host at least one speaker per semester and suggest other relevant technical discussions on the topics.

3. Mentoring