manageability


AMAX.AI Unveils [SMART]Rack Machine Learning Cluster - insideHPC

#artificialintelligence

Today AMAX.AI launched the [SMART]Rack AI Machine Learning cluster, an all-inclusive rackscale platform is maximized for performance featuring up to 96x NVIDIA Tesla P40, P100 or V100 GPU cards, providing well over 1 PetaFLOP of compute power per rack. The [SMART]Rack AI is revolutionary to Deep Learning data centers," said Dr. Rene Meyer, VP of Technology, AMAX. Solution components include: an All-Flash storage appliance for an ultra-fast in-rack data repository; 25G high-speed network; [SMART]DC HPC-optimized DCIM to remotely monitor, manage and orchestrate GPU-based Machine Learning hardware where real-time temperature, power and system health monitoring are critical for uninterrupted operation; and an in-rack battery for graceful shutdowns in the event of a power loss scenario. AMAX AI is a division of AMAX Information Technologies, Inc., specializing in award-winning high-performance platforms for AI and Deep Learning development.


Biologically Inspired Software Architecture for Deep Learning

#artificialintelligence

In the Google paper, the authors enumerate many risk factors, design patterns, and anti-patterns to needs to be taken into consideration in an architecture. These include design patterns such as: boundary erosion, entanglement, hidden feedback loops, undeclared consumers, data dependencies and changes in the external world. By contrast, Deep Learning systems (applies equally to machine learning), code is created from training data. A recent paper from the folks at Berkeley are exploring the requirements for building these new kinds of systems (see: "Real-Time Machine Learning: The Missing Pieces").