AI Service System

The AI-Service system – this is an IBM system built with the IBM Watson Machine Learning Accelerator (WMLA) AI platform. Users working on research studies in different disciplines can have access to the AI software/tools readily available on the WMLA platform for doing data preparation, model training and inference deployment for their AI/ML application developments. The hardware system resource, which provides the processing power of 1574 TFLOPS (FP16), can be summarized as follows:

  • 1 management node with 2 x IBM POWER9 CPU, 256GB RAM, 2 x 1.92TB SSD and 2 x NVIDIA V100 GPU

  • 6 AI compute nodes each with 8 x GeForce RTX GPUs (11GB), 2 x Intel Gold 6230R CPU, 384GB RAM, 32TB HDD and 1TB SSD

  • 2 AI compute nodes each with 4 x Titan RTX GPUs (24GB), 2 x Intel Gold 6230R CPU, 384GB RAM, 40TB HDD and 1TB SSD