How to Access tHe MRI-HPC (GPU) Cluster
This cluster was made possible by an NSF-MRI Grant received by Principal Investigators: Hongkai Yu , Zicheng Chi, Sathish Kumar, Mehdi Rahmati, Wenbing Zhao, Jacqueline Jenkins
Grant Details: NSF Link
Abstract
This project will establish a Graphics Processing Unit (GPU)-based High Performance Computing Instrumentation (HPCI) at Cleveland State University (CSU) to facilitate and promote Smart City research. The research goal is to transform Cleveland to a modern smart city with shorter commute time even during rush hours, ultra-low crime rate, a highly robust and secure electric power grid, and successful professional sports teams contending for championships. The requested major research instrumentation will enable research in the following five areas: 1) Smart Traffic; 2) Smart Vehicle; 3) Smart Internet of Things (IoT); 4) Smart Microgrid; and 5) Smart Sports. The Smart Traffic project will develop novel deep learning methods to extract useful information from remote sensing traffic images under complex urban environments. The Smart Vehicle project will improve the state-of-the-art in reliable communications and in data processing among 5G-enabled Connected Vehicles and the Vehicle-to-Everything. The Smart IoT project will develop novel machine learning methods for wireless data-driven communication and wireless intelligent sensing. The Smart Microgrid project will develop novel hybrid learning methods to ensure system resiliency, cost effectiveness, efficiency, and security of microgrid. The Smart Sports project will develop new computer vision methods to recognize and evaluate fine-grained player activities towards more efficient training and player evaluation. The research has the potential to transform Cleveland into an exemplary smart city, and the research outcome could be applicable to many other urban areas in the US. The requested HPCI will make a substantial improvement to the CSU high performance computing capabilities, which will positively attract potential collaborative research in the Cleveland Metropolitan Area and greatly improve the quality of research training at CSU. The requested HPCI will support the undergraduate/graduate student research, education and training in several computer science/engineering and civil engineering courses at CSU. The research supports the K-12 education of Cleveland Metropolitan Area and also supports the Hispanic-minority research and education. The website for the project is at https://engineering.csuohio.edu/mri-hpc/ The project website is maintained by the Computer Systems Specialist of the Washkewicz College of Engineering at Cleveland State University. The research outcome, user guidance, news, and project related information will be provided and updated in this website over the expected lifetime of the requested instrument.
Publications supported by this NSF MRI grant
1. Huiming Sun, Jin Ma, Qing Guo, Qin Zou, Shaoyue Song, Yuewei Lin, Hongkai Yu. Coarse-to-fine Task-driven Inpainting for Geoscience Images. IEEE Transactions on Circuits and Systems for Video Technology, 2023.
2. Jinlong Li, Runsheng Xu, Xinyu Liu, Jin Ma, Zicheng Chi, Jiaqi Ma, Hongkai Yu. Learning for Vehicle-to-Vehicle Cooperative Perception under Lossy Communication. IEEE Transactions on Intelligent Vehicles, 2023.
3. Jinlong Li, Runsheng Xu, Jin Ma, Qin Zou, Jiaqi Ma, Hongkai Yu. Domain Adaptive Object Detection for Autonomous Driving under Foggy Weather. IEEE Winter Conference on Applications of Computer Vision (WACV), 2023.
4. Xinyu Liu, Jinlong Li, Jin Ma, Huiming Sun, Zhigang Xu, Tianyun Zhang, Hongkai Yu. Deep Transfer Learning for Intelligent Vehicle Perception: a Survey. Green Energy and Intelligent Transportation, 2023.
Education related news
1. 03/03/2023, Hispanic AI Research Seminar between CSU and the University of Texas at El Paso.
2. 03/24/2023, Hispanic AI Research Seminar between CSU and the University of Texas at El Paso.
3. 10/13/2023, Cleveland State University and NSF MRI High Performance Computing Workshop. The related files and video record are publicized: [Workshop Schedule] [Project Overview by Hongkai Yu] [Cluster Usage Tutorial by Huiming Sun] [Video Record]
4. 10/13/2023, Cleveland State University and NSF MRI High Performance Computing Workshop’s Feedback: (1) 24/26 attendees agree that the knowledge or hirizon about Artificial Intelligence was improved by our workshop. (2) 24/26 attendees agree that the knowledge or hirizon of GPU based Parallel Computing was improved by our workshop. (3) 25/26 attendees agree that the knowledge about how to apply for the MRI GPU Cluster usage at CSU was improved by our workshop. (4) 25/26 attendees agree that the knowledge about how to use the MRI GPU Cluster was improved by our workshop. (5) 25/26 attendees are interested in somehow using the MRI GPU Cluster in the future. (6) 26/26 attendees agree that the MRI GPU Cluster will help the CSU research and education.
How to use this NSF MRI GPU Cluster for High Performance Computing at CSU?
This NSF MRI award supports a GPU Cluster with 40 NVIDIA A6000 GPU Cards for the High Performance Computing at CSU, in which each GPU card has 48GB memory for parallel computing tasks of Artificial Intelligence, Data Science, Simulation, etc.
Basic Requirement: The faculty, staff, PhD students, Postdocs, Graduate Research Assistants at CSU are eligible for the free usage of the MRI GPU Cluster. Because of some security reasons, you need a fixed lab computer located at CSU campus for this application and the laptop is not acceptable. This lab computer will be used for remote access of your account if this application is approved. One lab computer at CSU campus can only be used for one account.
1. Please carefully read the file [MRI Cluster-Management Plan.docx] before application.
2. Go to the [HPC MRI Access Request] and fill the online form. The CSU ID and the IP of a CSU campus lab computer is needed.
3. After your application is approved, you can read [MRI Cluster-Tutorial.docx] for detailed usages.