Reference

Glossary

---

High-Performance Computing (HPC) refers to the practice of using powerful computers, often in the form of HPC clusters, to solve complex computational problems that are beyond the capabilities of standard computers (ie, desktop, laptop computers or even workstations). HPC leverages its computational power often in the form of parallel processing to divide workloads into smaller tasks managed by the multiple CPU cores of a single node or distributed across numerous nodes. The use of parallel computing on supercomputers enables faster and more efficient execution of large, data-intensive computations.

Central Processing Unit (CPU), is the hardware component responsible for executing instructions and performing calculations. It’s the core of a computer system, managing and controlling all other components.

Graphics Processing Unit (GPU) is an electronic circuit originally designed to rapidly manipulate and alter memory to accelerate the creation of images in a frame buffer intended for output to a display device. GPUs are specialized processors optimized for handling complex visual data and parallel computations. While traditionally focused on graphics, GPUs are today increasingly used for high-performance computing tasks in areas like artificial intelligence, machine learning, and some scientific simulations such as molecular dynamics. The natural parallel capabilities of GPUs made them outperform CPUs in very specific domains of scientific computing.