Submission information
Submission Number: 14
Submission ID: 31
Submission UUID: 5ebd9200-9792-4fd9-8ad7-faf593436e22
Submission URI: /form/project
Created: Tue, 09/03/2019 - 13:02
Completed: Tue, 09/03/2019 - 13:02
Changed: Sat, 05/01/2021 - 19:44
Remote IP address: 130.215.55.243
Submitted by: Scott Valcourt
Language: English
Is draft: No
Webform: Project
Project Title | Student-Developed HPC Cluster for Active Learning |
---|---|
Program | Northeast |
Project Leader | Scott Valcourt |
sav@unh.edu | |
Mobile Phone | |
Work Phone | 6038624489 |
Mentor(s) | Scott Valcourt |
Student-facilitator(s) | Cristiano Bianchi |
Mentee(s) | |
Project Description | This project will involve the recommissioning of used computing hardware to craft a basic HPC cluster. Students will be solely responsible for the design, construction, configuration, and maintenance of a recommissioned collection of used computing hardware to craft a basic HPC cluster. The experiment will offer students practical experience in the administration and use of a HPC cluster, something not typically made available in the regular curriculum. Additionally, this project will serve as a potential cookbook to encourage other small and medium-sized institutions to reconsider scrapping old computing hardware for the purpose of teaching students HPC on a shoestring budget. |
Project Deliverables | Operating HPC cluster designed and built by students |
Project Deliverables | |
Student Research Computing Facilitator Profile | Ideally the students involved in this project are undergraduate or graduate students with experience in system administration of Linux/UNIX computers and have thought about HPC as a potential tool to be used in computing. |
Mentee Research Computing Profile | |
Student Facilitator Programming Skill Level | Some hands-on experience |
Mentee Programming Skill Level | |
Project Institution | University of New Hampshire |
Project Address | 268 Mast Road NHPBS Durham, New Hampshire. 03824 |
Anchor Institution | NE-University of New Hampshire |
Preferred Start Date | 07/09/2018 |
Start as soon as possible. | No |
Project Urgency | Already behind3Start date is flexible |
Expected Project Duration (in months) | |
Launch Presentation | |
Launch Presentation Date | |
Wrap Presentation | |
Wrap Presentation Date | |
Project Milestones | |
Github Contributions | |
Planned Portal Contributions (if any) | Paper and/or documentation on how to configure a HPC using recommissioned hardware and the design decisions required along the way as a workflow of sorts. |
Planned Publications (if any) | Paper on how a small institution can use recommissioned hardware to built a student-available HPC for real computation. |
What will the student learn? | HPC configuration decisions, system administration, cluster management |
What will the mentee learn? | |
What will the Cyberteam program learn from this project? | How to inexpensively provide assistance to small institutions in getting into HPC without having to spend a lot of money to try it out in-house. |
HPC resources needed to complete this project? | Old computing hardware, Science DMZ network access to connect to external resources |
Notes | |
What is the impact on the development of the principal discipline(s) of the project? | The development of a HPC asset that is part of the Eastern Regional Network (ERN) Proof of Concept National Research Platform (NRP) allowed for UNH to participate in something not available on campus, gave a student the chance to learn about HPC development which is not part of the standard curriculum, and connected HPC researchers to the regional community and envision a future asset that could benefit even more users over time. |
What is the impact on other disciplines? | Other disciplines have used the ERN environment to field test code and seek results. |
Is there an impact physical resources that form infrastructure? | The HPC was connected to the Eastern Regional Network (ERN) Proof of Concept asset across 14 institutions in the Northeast. |
Is there an impact on the development of human resources for research computing? | The opportunity for a student to learn about HPC system development and configuration was unique. |
Is there an impact on institutional resources that form infrastructure? | The asset created in this project is available for university researcher use and becomes another computing tool in computer science. |
Is there an impact on information resources that form infrastructure? | The outcomes report from this project will become a resource for future teams of students looking to build HPC resources and integrate them into the ERN or operate them in a stand-alone environment. |
Is there an impact on technology transfer? | There is no impact on technology transfer at this time. |
Is there an impact on society beyond science and technology? | Other than saving old hardware from the scrap yard, there isn't a large impact on society beyond science and technology. |
Lessons Learned | The student involved in this project learned that there are many aspects to developing HPC resources, but that there is a network of people already doing this work that are readily available to help answer questions and troubleshoot problems encountered. |
Overall results | The development of a HPC asset that is part of the Eastern Regional Network (ERN) Proof of Concept National Research Platform (NRP) allowed for UNH to participate in something not available on campus, gave a student the chance to learn about HPC development which is not part of the standard curriculum, and connected HPC researchers to the regional community and envision a future asset that could benefit even more users over time. |