Ceva Adds Homogenous AI Acceleration to Third-Gen Engine
By Sally Ward-Foxton, EETimes (January 13, 2022)
Ceva has revamped its NeuPro AI accelerator engine IP, adding specialized co-processors for Winograd transforms and sparsity operations and a general-purpose vector processing unit alongside the engine’s MAC array. The new generation engine, NeuPro-M, can boost performance 5-15X (depending on the exact workload) compared to Ceva’s second generation NeuPro-S core (released Sept 2019). For example, ResNet-50 performance was improved 4.9X without using the specialized engines – boosted to 14.3X when using specialized co-processors, according to Ceva. Results for Yolo-v3 showed similar speedups. The core’s power efficiency is expected to be 24 TOPS/Watt for 1.25 GHz operation.
The NeuPro-M engine architecture allows for parallel processing on two levels — between the engines (if multiple engines are used), and within the engines themselves. The main MAC array has 4000 MACs capable of mixed precision operation (2-16 bits). Alongside this are new, specialized co-processors for some AI tasks. Local memory in each engine breaks the dependence on the core shared memory and on external DDR; the co-processors in each engine can work in parallel on the same memory, though they sometimes transfer data from one to another directly (without passing through memory). The size of this local memory is configurable based on network size, input image size, number of engines in the design and customers’ DDR latency and bandwidth.
E-mail This Article | Printer-Friendly Page |
|
Ceva, Inc. Hot IP
Related News
- AImotive's latest aiWare3P delivers superior NN acceleration for production L2-L3 automotive AI
- Ceva Launches Multi-Protocol Wireless Platform IP Family to Accelerate Enhanced Connectivity in MCUs and SOCs for IoT and Smart Edge AI Applications
- Blumind Harnesses Analog for Ultra Low Power Intelligence
- Kalray and Arm to collaborate to bring data intensive processing and AI acceleration DPU solutions to the global Arm ecosystem
- Ceva Expands AI Ecosystem for its Class-Leading NPU IP with New Partnerships for Automotive and Edge AI
Breaking News
- JEDEC Updates JESD79-5C DDR5 SDRAM Standard: Elevating Performance and Security for Next-Gen Technologies
- Alphawave Semi: FY 2023 and 2024 YTD Trading Update and Notice of Results
- Silvaco Announces Expanded Partnership with Micron Technology
- U.S. Subsidy for TSMC Has AI Chips, Tech Leadership in Sight
- Zhuhai Chuangfeixin: OTP IP Based on 90nm CMOS Image Sensor Process Technology Successfully Mass Production
Most Popular
- Semiconductor Capacity Is Up, But Mind the Talent Gap
- Microchip Technology Acquires Neuronix AI Labs
- CMC Microsystems and AIoT Canada Sign Memorandum of Understanding to support IoT and semiconductor ecosystem growth in Canada
- Brazil and Europe sign innovative project with RISC-V technology for HPC
- Alphawave Semi - Q4 2023 Trading and Business Update