Computing is changing - energy constraints and performance bottlenecks create new opportunities for unique and novel hardware designs. However, these new systems require new approaches for application development, management, and visualization. Our main goal for this project is to help interface new “post-Moore” hardware like neuromorphic field-programmable analog arrays (FPAAs), memory-centric architectures like the Emu Chick, and tightly coupled FPGAs with current software technologies. We aim to develop easy-to-use software demonstrations and examples for each new piece of hardware in the Rogues Gallery testbed and to design strategies to extend the hardware for interesting research problems and class projects.
Issues Involved or Addressed
Extremely heterogeneous systems management, neuromorphic and analog design, software tools for novel hardware, processing of sensitive / encrypted data.
Methods and Technologies
Academic Majors of Interest
Preferred Interests and Preparation
Required skills include: Interest in and willingness to learn basic Linux commands, experience with either Python or C programming. Willingness to learn/use Markdown and git.
Suggested skills include:
CompE: Background/interest in FPGA or FPAA programming; Completion of ECE 3400 would be helpful
CS/CSE: Background/interest in high performance computing, encryption of data, data analysis, python GUI and python-related analysis tools.
Industrial Engineering: Background/interest in workflow design, machine learning
Mechanical Engineering: Interest in sensors, power monitoring, and embedded sensor design.
Computational Media: Background/interest in interactive GUI design, visualization, and or web design.