Gregor von Laszewski, Ph.D. https://laszewski.github.io/ gvonlasz@iu.edu, laszewski@gmail.com
Supervisor Qualifications
Dr. von Laszewski has a track record supporting undergraduates at any organizations he taught. Examples of past successes include an IU award for the best undergraduate project, a Novell developers award, an RIT award for an outstanding undergraduate student research project, the Department of Energy outstanding mentor award, and IU staff award.
Research
In this project, we will be exploring the use of Raspberry Pis for Data Science. Each student will have a minimum of five Raspberry PIs to build a research cluster. The students will explore setting up a Kubernetes, Hadoop, Spark, Kubernetes, or compute services using REST in a reproducible fashion. An application can be chosen based on the students' interest in data science. Examples of applications would be image detection of features recorded on an autonomous robot boat or COVID-19 time series data analysis. If successful, the students can build a cluster with more than 100 Raspberry Pi's at a Lab and test their application for benchmarking comparisons.
Deliverables
- Successful creation of a raspberry pi cluster with the newest software for the PI
- Documentation of the setup in a manual so others can replicate the setup
- Academic Paper describing the project based on the interest and focus of the student
- Weekly updates
- Having fun while doing research
Links
- https://github.com/cloudmesh/cloudmesh-pi-burn
- https://github.com/cloudmesh/cloudmesh-pi-burn/blob/main/paper/vonLaszewski-cmburn.md
- https://github.com/cloudmesh/cloudmesh-pi-cluster
- We have more information, but this gets you started
- you can immediately engage with Gregor at laszewski@gmail.com
Results from Previous UROCs
 Research Paper |
 Poster |
Prerequisites
- Knowledge of Python is required. We will use elementary Python, including variables, conditions, loops, functions, classes, modules.
- Elementary Knowledge of Linux command line is highly recommended. You will need to know how to manage directories, create files, edit them. Typically this can be learned in less than a day. If you do not know it, you will be obtaining it as part of the research project.
Equipment
This project will follow the IU regulations on COVID-19 social distancing for Labs. Furthermore, The student will be given a minimum of five Raspberry Pis, one or more network hubs, cables, SD-cards so the project can be performed at the student's residence.
 |
 |
| Fig 1: Raspberry Pi cluster Prototype | Fig 2: Configuration we will setup with our exiting easy to use tools |