CyberInfrastructure Description Repository

PSC Bridges-2 GPU-AI (Bridges-2 GPU Artificial Intelligence)

Pittsburgh Supercomputing Center

Production Dates: 02/11/2021 - 09/30/2025

Public URL: https://cider.access-ci.org/public/resources/RDR_000705



Description: Bridges-2 Accelerated GPU (GPU) nodes are optimized for scalable artificial intelligence (AI; deep learning). Bridges-2 GPU nodes each contain 8 NVIDIA Tesla V100-32GB SXM2 GPUs, providing 40,960 CUDA cores and 5,120 tensor cores. In addition, each node holds 2 Intel Xeon Gold 6248 CPUs; 512GB of DDR4-2933 RAM; and 7.68TB NVMe SSD. They are connected to Bridges-2's other compute nodes and its Ocean parallel filesystem and archive by two HDR-200 InfiniBand links, providing 400Gbps of bandwidth to enhance scalability of deep learning training. Bridges-2 GPU AI resource will also contain 9 HPE 8-Volta servers along with a NVIDIA DGX-2.

Recommended Use: Bridges-2 GPU nodes are optimized for scalable artificial intelligence (AI) including deep learning training, deep reinforcement learning, and generative techniques. Specific architectural features that support this are eight fully interconnected V100-32GB SXM2 GPUs per node and dual-rail HDR-200 between nodes, which together support larger models and scaling across multiple nodes. Bridges-2 GPU nodes are also applicable to accelerated simulation and modeling. Applications using Bridges-2 GPU should make use of tensor cores to realize reasonable efficiency.

User Guide: https://portal.xsede.org/psc-bridges-2

Gateway Recommended Use: Bridges-2 GPU nodes are available as back-end computational engines. Bridges-2 also contains dedicated RM-type nodes for web servers and persistent databases.