Supercomputing Industry HPC Cluster Introduction

Feb 17, 2023


What High-Performence Applications and Needs Are There in Supercomputing?

In the field of supercomputing, powerful computers are used to solve complex problems that require a large amount of computing power. Supercomputers in HPC data centers are used for tasks such as weather forecasting, scientific research, and data analysis, etc. Nowadays supercomputing capability has been widely used in many fields:

  1. Weather and Climate
  2. Materials Science
  3. Wind Tunnel Simulation
  4. Genomics
  5. Physics
  6. Computational Chemistry
  7. Data Science
  8. Quantum Chemistry

What Software Stacks Are Typically Used in Supercomputing HPC Cluster?

The software stack used in supercomputing HPC clusters typically includes operating systems such as Linux, programming languages ​​such as C/C++ and Fortran, and libraries such as MPI and OpenMP. Additionally, some specialized software is used for AI-accelerated high-precision HPC clusters in distributed computing, such as Cryosparc, FUN3D, etc.

  • Cryosparc
  • FUN3D
  • Microevolution
  • Clara
  • WRF

What Hardware Are Typically Used in Supercomputing HPC Cluster?

Besides the high-performance processors, large amounts of RAM, and high-performance storage systems used in HPC clusters from vendors like Intel, Dell, AMD, NVIDIA, HPE, Supermicro, etc., many supercomputing HPC clusters use specialized hardware such as GPU servers and GPUs, FPGAs, high-speed interconnect network adapters, switches and HPC cables (Mellanox was a common and usually the only pick in the past) to build accelerated computing in the data center infrastructure.

Need Double Precision Computing (FP64)?
Recommended GPU Model: NVIDIA A100/A30.

Need Low Latency Access to Storage, and Supports for GPUDirect RDMA?
Use NVIDIA Mellanox InfiniBand NICs.

Need Connectivity Assemblies to Build Low-Latency and High-Performance InfiniBand Cluster?
Use 200Gb/s HDR InfiniBand active optical cables/passive copper cables/transceivers.

Related Resources:
Case Study: NADDOD Helped the National Supercomputing Center to Build a General-Purpose Test Platform for HPC