HPC Cluster Deployment Project

This project aims to deploy a High-Performance Computing (HPC) cluster at a fictive university.

Project Goals

• Deploy a high-performance, scalable HPC cluster
• Provide a robust and reliable platform for research and teaching
• Meet the needs of the university's research community
• Support a wide range of scientific and engineering applications

Key Components

• Compute Nodes: nodes for parallel computing tasks.
• Login Nodes: nodes for users to access and manage the cluster.
• Master Nodes: nodes responsible for job scheduling and cluster management.
• Service Nodes: nodes for additional services like file storage.
• Router Nodes: nodes for high-speed communication between cluster components.
• Network: Infiniband HDR200 for fast and efficient data transfer.
• Storage: Lustre parallel file system for large datasets.

Software:

• Red Hat Enterprise Linux 9 (RHEL9) operating system
• Slurm job scheduler for managing job submissions
• GNU and Intel compiler suites for scientific computing
• NVIDIA CUDA toolkit for GPU-accelerated applications
• MATLAB and MATLAB-Engine for mathematical computations

Project Deliverables

• A fully functional HPC cluster
• Documentation for installation and configuration
• User training on how to utilize the cluster effectively