Corerain’s Custom AI Streaming Accelerator (CAISA) Architecture 3.0 is a high-performance computing architecture designed for deep learning neural networks inference.
The CAISA architecture controls the computing sequence through the sequence of the data streams. Operator-level data flow graphs are analyzed from the CNN network model and then mapped to the CAISA streaming engine. It eliminates idle cycles by overlapping data movement and data calculation. By maximizing chip utilization ratio, it provides cost effective advanced computing power for AI applications.
CAISA 3.0 and RainBuilder support Corerain products with high performance and high expandability.
High Cost Efficiency
Cycle accurate computation scheduling
Streaming network dynamic reconfiguration
Automated end-to-end deployment
Shenzhen | 14F, Changfu Jinmao Building(CFC), Trade-free Zone, Futian District, Shenzhen, Guangdong, China
UK | Kemp House,152-160 City Road, London, EC1V 2NX
Shanghai | Room C509, 5F, Building C, Hongqiao Wantong Center, No.333 Su Hong Road, Minhang District, Shanghai, China
Jinan | 14F, Building 3, Future Innovation Park, 3 Gangxing Road, Licheng District, Jinan, Shandong，China
Xi 'an | Room 1211, 12F, Building T1, Lai 'an Center, 1111 Yanta Road, Qujiang New District, Xi 'an, Shaanxi , China
Shenyang | Unit 064A, 6F, Huahang Building, 77 Wenhua Road, Heping District, Shenyang, Liaoning , China