Partitioning to optimize AI inference for multi-core platforms
By Rami Drucker, Ceva
EDN (January 8, 2024)
Not so long ago, artificial intelligence (AI) inference at the edge was a novelty easily supported by a single neural processing unit (NPU) IP accelerator embedded in the edge device. Expectations have accelerated rapidly since then. Now we want embedded AI inference to handle multiple cameras, complex scene segmentation, voice recognition with intelligent noise suppression, fusion between multiple sensors, and now very large and complex generative AI models.
Such applications can deliver acceptable throughput for edge products only when run on multi-core AI processors. NPU IP accelerators are already available to meet this need, extending to eight or more parallel cores and able to handle multiple inference tasks in parallel. But how should you partition expected AI inference workloads for your product to take maximum advantage of all that horsepower?
To read the full article, click here
Related Semiconductor IP
- Multi-channel, multi-rate Ethernet aggregator - 10G to 400G AX (e.g., AI)
- Multi-channel, multi-rate Ethernet aggregator - 10G to 800G DX
- 200G/400G/800G Ethernet PCS/FEC
- 50G/100G MAC/PCS/FEC
- 25G/10G/SGMII/ 1000BASE-X PCS and MAC
Related White Papers
- AI Edge Inference is Totally Different to Data Center
- Why Interlaken is a great choice for architecting chip to chip communications in AI chips
- The Growing Importance of AI Inference and the Implications for Memory Technology
- Top 5 Reasons why CPU is the Best Processor for AI Inference
Latest White Papers
- e-GPU: An Open-Source and Configurable RISC-V Graphic Processing Unit for TinyAI Applications
- How to design secure SoCs, Part II: Key Management
- Seven Key Advantages of Implementing eFPGA with Soft IP vs. Hard IP
- Hardware vs. Software Implementation of Warp-Level Features in Vortex RISC-V GPU
- Data Movement Is the Energy Bottleneck of Today’s SoCs