The Dell-EMC DDS 8440 IPU Machine Learning Server is a 4U rack-mounted server, containing eight Graphcore C2 IPU-Processor PCIe Cards, each of which has two Graphcore Colossus GC2 IPUs (Intelligence Processing Units), supported by Graphcore’s Poplar software stack. Designed to support both training and inference on the same hardware for full utilization of your machine intelligence compute, the DSS8440 IPU Server is a new tool in your machine intelligence toolkit.
The DSS 8440 is a 4U 2-socket accelerator-optimized server designed to deliver exceptionally high compute performance for both training and inference. Its open architecture, based on a high performance switched PCIe fabric, maximizes customer choice for machine learning infrastructure while also delivering best of breed technology. It lets you tailor your machine learning infrastructure to your specific needs – with open, PCIe based components.
The DSS 8440 is specifically designed to reduce time to insight for machine learning training by providing substantially increased horsepower. Analyze massive amounts of data, recognize patterns, and determine follow-on actions, with up to 10 Nvidia GPU's or 8 Graphcore C2 IPU-Processor PCIe Cards.
Its high speed, switched PCIe fabric and extensive local storage (NVMe & SATA) contribute to achieving rapid results. The Graphcore C2 IPU Card is a purpose built processor chip designed specifically to run machine learning workloads.
The DSS 8440 IPU delivers leading training performance and efficiency (performance/watt) using the most common frameworks and popular convolutional neural network models. Further details can be found here.