英伟达 CUDA® 内核具有 GPU 加速功能，可支持通用计算。凌华科技的 GPU 解决方案与 Tensor Cores 相结合以加速 AI 推理和训练，此外还与 RT Cores 组合，采用实时光线追踪来提升渲染性能和速度，提供电影般影像，从而以更快的速度提供更好的结果。
凌华科技的 MXM GPU 模块尺寸仅为 PCIe 显卡的五分之一，可最大限度地减少主机系统的占用空间。这些模块经过专门的设计，能够承受工业环境的恶劣实际挑战，包括振动、冲击和极端温度。凌华科技的 MXM GPU 模块也很省电，最低功耗仅 20 瓦。
There is a growing need for GPUs at the edge. Compared with traditional graphics cards, ADLINK embedded MXM graphics cards offer similar performance, but is more power-efficient and only one-fifth the size. Also, they can survive severe temperature extremes, shock and vibration. ADLINK embedded MXM graphics cards can deliver AI inference and compute acceleration to size, weight and power constrained edge applications.
The need for computing is growing exponentially at the edge. The market size is expected to reach USD 155.9 billion by 2030, expanding at a CAGR of 38.9%*. Leveraging the power of GPUs enables industries to better position for edge computing. Download the infographic to learn seven ways to facilitate the edge transition with embedded MXM GPU modules.
Bringing AI to the edge provides many benefits, including faster response time, enhanced security, improved mobility, and lower communication costs. As combinations of neural networks and frameworks, running on specialized computing cores, are ideal for specific tasks, heterogeneous computing is the best strategy for deploying AI. Download the solution brief to learn how to optimize an AI platform with CPU, GPU, FPGA, and ASIC.