GAUDI vs Intel Nervana Neural Network Processor comparison

Gaudi is a processor that handles machine learning training workloads, and has 32GB of memory built-in, a memory bandwidth of 1TB per second, and consumes up to 200W. The only AI processor with Integrated RDMA over Converged Ethernet to provide scalability and lower total cost of ownership. Gaudi is designed for versatile and efficient system scale out and scale up with integrated on-chip RoCE RDMA, enabling high-performance inter connectivity.

The Intel Nervana Neural Network Processor (NNP) is a purpose built architecture for deep learning. This new architecture provides the needed flexibility to support all deep learning primitives while making core hardware components as efficient as possible.

Deep learning is taking the next leap forward. Increasingly sophisticated and complex data, models, and techniques will allow AI to move beyond identifying information to understanding context – enabling a level of “common sense” for reasoning and decision-making.