Wave Computing’s Native Dataflow Technology
Wave Computing’s unique approach to accelerating deep learning is based upon harnessing dataflow technology to eliminate the need for a host and co-processor in the processing of a neural network. This enables the company’s WaveFlow™ deep learning systems to exploit data and model parallelisms present in convolutional and recurrent neural networks to provide high-performance, high-efficiency training and inferencing computing solutions that scale for any implementation.
Wave Computing’s dataflow-based systems for the datacenter and on-premise environments each include Dataflow Processing Units (DPUs) that contain more than 16,000 Processing Elements per chip, high-speed memories, terabytes of storage and the company’s full dataflow software stack: the WaveFlow SDK, the WaveFlow Agent Library, WaveFlow Execution Engine, and the Wave Machine Learning Framework Interface.
The benefits of Wave Computing’s dataflow-based solutions include fast and easy neural network development and deployment using frameworks such as ONNX, TensorFlow, Keras and more. We invite you to learn more about our dataflow technology, which has been presented at industry-leading technical conferences including Hot Chips, IEEE ICCAD, the Processor Forum and more.
A Coarse Grain Reconfigurable Array (CGRA) for Statically Scheduled Data Flow Computing
to a non-VonNeumann
Designed to Scale