WhisperStation™- Deep Learning – EOL
Ultra-Quiet Computing for Deep Learning Researchers
Note: NVIDIA has reached the end of their production runs for Titan RTX and RTX 2080Ti GPUs. Microway’s custom-liquid cooled configuration builds with these GPUS have therefore been suspended, but our team can still design you a custom configuration for Deep Learning/AI Training around the latest GPUs in an air cooled design. Please contact us for help in designing such a system.
Designed for deep learning research, this WhisperStation features a high-speed Intel Xeon processor, up to 512GB DDR4 memory, four NVIDIA RTX 2080Ti or Titan RTX* GPUs, high-speed NVMe SSDs for the operating system, large-capacity storage, and ultra-quiet fans. This system is available with liquid cooling for the ultimate performance.
We provide a turn-key system delivered with Ubuntu Linux, NVIDIA CUDA, NVIDIA cuDNN and your choice of AI containers pre-installed from the NVIDIA GPU Container registry.
WhisperStation-Deep Learning has been designed to provide maximum performance when training deep neural networks. It is the perfect system for a researcher exploring the applications of machine learning before such applications move into production. For production-level deep learning, we recommend our professional NumberSmasher server product line with NVIDIA Datacenter GPUs.
With NGC containers, users can realize the latest software optimization and performance-direct from NVIDIA- for their workload. Each container includes a fully optimized framework, and users can train their networks with incredible performance.