Skip to content

HPC Education and Training

NJIT HPC provides practical training in high performance computing for students and researchers at various levels of expertise. HPC training for research professionals aims to enhance their capabilities in utilizing high-performance computing, data-intensive computing, and data analytics within their respective research fields.

Training Programs

2024 Fall

  • Job Arrays and Advanced Submission Techniques for HPC

    This is the final in a series of three webinars in the fall semester. designed to introduce researchers, scientists, and HPC users to the fundamentals of the containers. This session aims to provide useful information on submitting SLURM jobs efficiently by covering job arrays, job dependencies, checkpointing, and addressing common SLURM job issues.

    Key Highlights:

    • Understanding the concept and benefits of job arrays
    • Syntax for submitting and managing job arrays
    • Best practices for efficient array job design
    • Dependency chains and complex workflows
    • Resource optimization strategies
    • Using SLURM's advanced options for improved job control
    • Checkpointing the jobs and use of 3rd party checkpointing tool

    Download Slides

  • Introduction to Containers on Wulver

    This is the second in a series of three webinars in the fall semester, designed to introduce researchers, scientists, and HPC users to the fundamentals of the containers. Attendees will learn the fundamentals of Singularity, including installation, basic commands, and workflow, as well as how to create and build containers using definition files and existing Docker images. The training will cover executing containerized applications on HPC clusters and integrating with job schedulers like SLURM, while also addressing security considerations and performance optimization techniques.

    Key Highlights:

    • Introduction to containers and its role in HPC environments
    • Fundamentals of Singularity, including installation, basic commands, and workflow
    • Create and build containers using definition files and existing Docker images
    • How to execute containerized applications on HPC clusters
    • Use Containers via SLURM
    • Performance optimization techniques

    Download Slides

  • SLURM Batch System Basics

    This is the first in a series of three webinars in the fall semester. designed to introduce researchers, scientists, and HPC users to the fundamentals of the SLURM (Simple Linux Utility for Resource Management) workload manager. This virtual session will equip you with essential skills to effectively utilize HPC resources through SLURM.

    Key Highlights:

    • Introduction to SLURM and its role in HPC environments
    • Basic SLURM commands for job submission, monitoring, and management
    • How to write effective job scripts for various application types
    • Understanding SLURM partitions, quality of service, and job priorities
    • Best practices for resource requests and job optimization
    • Troubleshooting common issues in job submission and execution

    Download Slides

2024 Spring

Since Wulver is quite different from the older cluster Lochness, the HPC training programs are designed to guide both new and existing users on how to use the new cluster. The following trainings provide the basic information on

  • Introduction to HPC
  • Performance Optimization
  • Job Submission and Management
  • Managing Conda Environment

If you still have any questions on HPC usage, please contact the HPC Facilitator.

  • Getting Started on Wulver: Session I

    This is the first in a series of three webinars introducing the NJIT HPC environment. This webinar provided the basic information in learning more about our new High Performance Computing (HPC) research cluster, Wulver.

    Key Highlights:

    • HPC concepts
    • Hardware and architecture of the Wulver cluster
    • Guidance on how to obtain an account and receive an allocation to utilize the shared resources.

    Download Slides

  • Getting Started on Wulver: Session II

    This session offered an overview of the environment on the Wulver cluster, including file management, working with the batch system (SLURM), and accessing software.

    Key Highlights:

    • HPC allocations
    • Using SLURM
    • Job submissions

    Download Slides

  • Introduction to Python and Conda

    Participants will gain an introductory understanding of using Python for HPC and effectively managing their Python environments using Conda. This knowledge will empower them to leverage the power of Python for their scientific computing needs on HPC systems.

    Key Highlights:

    • Learn how to manage Python environments for HPC using Conda.
    • Become familiar with common tools and libraries for scientific computing in Python.

    Download Slides