Fornitore: Firefly
CSA1-N8S1684X AI Computing Power 1U Server
Descrizione
The CSA1-N8S1684X server is equipped with the high-computing BM1684X, which can be configured with up to 8 BM1684X computing modules, providing excellent computing power, up to 256TOPS (INT8) peak computing power, or 128TFLOPS (FP16/BF16) computing power, or 16TFLOPS (FP32) high-precision computing power, with powerful multi-channel video processing capabilities. It provides rich algorithm transplantation and a one-stop deep learning development toolkit, and supports secondary development with the BMC management system.
- Product Datasheet: Click Here
- Product Wiki: Click Here
Feature
- AI Processor BM1684X: The BM1684X AI computing SoC chip adopts an octa-core Cortex-A53 architecture + NPU architecture. The Cortex-A53 has a main frequency of 2.3GHz, and the NPU provides a computing power of 32Tops (INT8) / 16TFLOPS (FP16/BF6) / 2TFLOPS (FP32).
- Eight Computing Modules: The server can be equipped with up to 8 BM1684X computing modules. Users can customize the number of modules and storage configuration, making it flexible for various scenarios.
- Multi-channel Video Processing Capability: The server supports up to 256 channels of H.264 1080p@25fps video decoding, 256 channels of H.264 1080P@25fps HD video processing (decoding + AI analysis), and 96 channels of H.264 1080p@25fps video encoding.
- Diverse Algorithms and Strong Practicality: The server supports the migration of algorithms such as pedestrian/vehicle/object recognition, video structuring, and trajectory behavior analysis. It has high security and reliability and can be flexibly applied to the development of various products.
