With the support of Nvidia, IBM has unveiled a new storage system that has been constructed for artificial intelligence (AI) workloads. The instrument is designed to be used with information tools such as TensorFlow Spark or PyTorch. It’s a system blending all-flash drives, file systems and the Nvidia DGX-1 GPU.
The essentials of the system include IBM Spectrum Scale v5 file storage DGX-1 servers, the Nvidia DGX software heap optimized for training functionality; NVMe storage per 2U block with 300TB; and data throughput.
The offering is available through Nvidia and IBM, in addition to their resellers. It’s designed to work with IBM spectrum Scale. In accordance with Storage Review, Nvidia is ‘impressively well suited’ for AI research with its DGX-1.
“The system has eight Tesla V100 GPUs that provide a total of 256GB GPU memory. The CPU is typically a dual-core Intel Xeon. With 40,960 CUDA cores and 5,120 Tensor cores, this beast will be able to handle any AI model you throw at it”
Recently, storage vendors have been consolidating ranks and focusing harder on all-flash architectures, better designed for artificial intelligence and machine learning workloads. Nvidia has been putting in some work, teaming up with Actual Storage to attain that goal. NetApp, on the other hand, partnered for the identical purpose with Lenovo.