CEVA unveils NeuPro-M heterogeneous processor architecture

2 min read

CEVA has announced NeuPro-M, its latest generation processor architecture for artificial intelligence and machine learning (AI/ML) inference workloads.

Intended for the Edge AI and Edge Compute markets, NeuPro-M is a self-contained heterogeneous architecture that is composed of multiple specialized co-processors and configurable hardware accelerators that process diverse workloads of Deep Neural Networks, boosting performance by 5-15X compared to its predecessor.

In what is claimed to be an industry first, NeuPro-M supports both system-on-chip (SoC) as well as Heterogeneous SoC (HSoC) scalability to achieve up to 1,200 TOPS and offers optional robust secure boot and end-to-end data privacy.

NeuPro–M compliant processors initially include the following pre-configured cores:
  • NPM11 – single NeuPro-M engine, up to 20 TOPS at 1.25GH
  • NPM18 – eight NeuPro-M engines, up to 160 TOPS at 1.25GHz

According to CEVA, a single NPM11 core, when processing a ResNet50 convolutional neural network, achieves a 5X performance increase and 6X memory bandwidth reduction versus its predecessor, which results in power efficiency of up to 24 TOPS per watt.

The NeuPro-M is capable of processing all known neural network architectures, as well as integrated native support for next-generation networks like transformers, 3D convolution, self-attention and all types of recurrent neural networks. NeuPro-M has been optimised to process more than 250 neural networks, more than 450 AI kernels and more than 50 algorithms.

The embedded vector processing unit (VPU) ensures future proof software-based support of new neural network topologies and new advances in AI workloads. Furthermore, the CDNN offline compression tool can increase the FPS/Watt of the NeuPro-M by a factor of 5-10X for common benchmarks, with very minimal impact on accuracy.

The NeuPro-M heterogenic architecture is composed of function-specific co-processors and load balancing mechanisms that are the main contributors to the huge leap in performance and efficiency compared to its predecessor.

By distributing control functions to local controllers and implementing local memory resources in a hierarchical manner, the NeuPro-M achieves data flow flexibility that result in more than 90% utilization and protects against data starvation of the different co-processors and accelerators at any given time. The optimal load balancing is obtained by practicing various data flow schemes that are adopted to the specific network, the desired bandwidth, the available memory and the target performance, by the CDNN framework.

NeuPro-M architecture highlights include:

  • Main grid array consisting of 4K MACs (Multiply And Accumulates), with mixed precision of 2-16 bits
  • Winograd transform engine for weights and activations, reducing convolution time by 2X and allowing 8-bit convolution processing with <0.5% precision degradation
  • Sparsity engine to avoid operations with zero-value weights or activations per layer, for up to 4X performance gain, while reducing memory bandwidth and power consumption
  • Fully programmable Vector Processing Unit, for handling new unsupported neural network architectures with all data types, from 32-bit Floating Point down to 2-bit Binary Neural Networks (BNN)
  • Configurable Weight and Data compression down to 2-bits while storing to memory, and real-time decompression upon reading, for reduced memory bandwidth
  • Dynamically configured two level memory architecture to minimize power consumption attributed to data transfers to and from an external SDRAM
The NeuPro-M architecture also supports secure access in the form of optional root of trust, authentication, and cryptographic accelerators.

NeuPro-M provides a fully programmable hardware/software AI development environment for customers to maximize their AI performance. CDNN includes innovative software that can fully utilize the customers’ NeuPro-M customised hardware to optimise power, performance & bandwidth.

The CDNN software also includes a memory manager for memory reduction and optimal load balancing algorithms, and wide support of various network formats including ONNX, Caffe, TensorFlow, TensorFlow Lite, Pytorch and more.

CDNN is compatible with common open-source frameworks, including Glow, tvm, Halide and TensorFlow and includes model optimisation features like ‘layer fusion’ and ‘post training quantization’ all while using precision conservation methods.

NeuPro-M is now available for licensing to lead customers and for general licensing in Q2 this year