Student Research Opportunities

My research falls into multiple areas, as I typically collaborate with professors in CSE, CS, and ECE (Drs. Vuduc, Kim, Conte, and Sarkar). However, I generally would classify my work at the intersection of high-performance computing and computer architecture, meaning that I am interested in algorithmic and architecture solutions to a variety of problems.

Course credit is available and sufficient skills/experience may lead to a paid position. If you are interested, please check out the projects below and upload your resume with related skill set and experience at this GT survey link: Research Interest Survey

Slides for 2022 Research Opportunities: UROC slides

VIP Class Link

What’s the goal? This VIP class provides an introduction to novel computing paradigms including quantum, neuromorphic, reconfigurable, and near-memory computing. Students are put into subteams and they learn the skills to tackle a research problem for a specific topic area in the space of “post-Moore” computing.

Required skills: Some Python or C++ experience and an interest to learn more about novel computing paradigms!

Student Receives: 1,2, or 3 credits

Future opportunities: Many VIP students go on to do paid research or research for credit in research groups working on related topics.

SuperSTARLU Research for Undergrads (REU)

Traditional High-performance computing (HPC) is primarily focused on running large, scientific codes in parallel across large computer clusters called supercomputers. This project looks at a specfic linear solver application, SuperLU, and attempts to map it to an FPGA platform with a “stacked” memory. Additionally, we are interested to profile sparse access patterns from this application and related applications in the graph analytics space.

What’s the goal? We are looking to develop and understand a “memory-aware” version of the SuperLU linear solver and related codes.

Required skills A basic understanding of using Linux, ssh, and accessing X-based software remotely as well as some general experience with programming in C or C++.

Desired skills Experience with OpenMP or having taken one of the HPC or parallel programming classes (CX 4220) or experience with computer architecture simulators (CS 4290). Experience with FPGA development and related courses (CS 3220).

Student Receives: Paid position if skill set and results fit longer-term project. $10-12/hour.

Neuromorphic Computing Platform Development

Neuromorphic or brain-inspired computing may provide a low-power means to perform machine learning computations

What’s the goal? This research aims to tie together an existing neuromorphic hardware platform, the Field Programmable Analog Array, and software techniques for programming it.

Required skills A basic understanding of using Linux, ssh, and accessing X-based software remotely as well as some general experience with programming in C or C++.

Desired skills: Experience with machine learning concepts. Experience with using the FPAA as part of the ECE curriculum (ECE 3400/4430/6435).

Student Receives: Credit for undergraduate or graduate research.

Future opportunities: This work ties in with the goals of our CRNCH Rogues Gallery testbed and is also of relevance to opportunities at GTRI and at labs at Georgia Tech and Oak Ridge National Laboratory.

The CRNCH Rogues Gallery testbed is looking for students to help develop the next generation of novel computing.

What’s the goal? This research and development effort is focused on building the testing and training infrastructure for the Rogues Gallery testbed. Students will help to test and develop new training material for deployment in the testbed.

Required skills A basic understanding of using Linux, ssh, and accessing X-based software remotely as well as some general experience with programming in C or C++.

Desired skills: Experience with Linux systems and familiarity with any of the following: Jupyter notebooks, Docker, Singularity, Slurm, Vivado, Salt/Ansible.

Student Receives: Credit or pay; pay is $10-12/hour based on experience.

Quantum Computing Algorithm Development

Quantum computing software still requires more effort to determine the future “killer applications” that will make quantum systems useful as hybrid accelerators.

What’s the goal? This research aims to accelerate standard high-performance computing kernels with quantum algorithms for the QCOR quantum programming language.

Required skills A basic understanding of using Linux, ssh, and linear algebra. Some general experience with programming in C or C++.

Desired skills: Experience with quantum computing frameworks like Qiskit, Q#, or QCOR.

Student Receives: Credit or pay; pay is $10-12/hour based on experience.