To build advanced high-performance computing (HPC) applications efficiently, it’s critical to leverage HPC container technologies like Docker and Singularity to access many application components – libraries, software stacks, and more – in a way that is portable and easy to reproduce.
In the new NVIDIA Deep Learning Institute (DLI) online course on High-Performance Computing with Containers, you’ll learn how to reduce complexity and improve the portability of your code using containerized environments for HPC application development.
Through hands-on, self-paced learning powered by GPU-accelerated workstations in the cloud, you’ll:
- Explore the basics of building and running Docker and Singularity containers
- Leverage the HPC Container Maker (HPCCM) to programmatically configure complex, portable, bare-metal HPC environments for your application
- Apply advanced container building techniques like layered containers and multi-stage builds
- Use drop-in containerized versions of existing HPC applications like MPI Bandwidth and MILC
At the end of the two-hour course, you’ll be able to quickly build and use Docker, Singularity, and HPCCM for portable, bare-metal performance in your HPC applications.
To get started, enroll in the course and create or log into your NVIDIA Developer Program account. All you need is your laptop and internet connection.
Want more hands-on experience building HPC applications? Earn a certificate in the fundamentals of accelerated computing for CUDA C/C++, CUDA Python, and OpenACC by enrolling in an eight-hour, self-paced course online. Or, check out our instructor-led workshops for accelerated computing, designed for teams of developers and led by DLI-certified instructors. For more information, visit www.nvidia.com/dli.