NVIDIA rules the performance roost, Qualcomm demonstrates exceptional power efficiency, and Intel demonstrates the power of software. This time around NVIDIA, Intel, and Qualcomm added new models and configurations, and results were submitted from Dell, HPE, Lenovo, NVIDIA, Inspur, Gigbyte, Supermicro, and Netrix. On that front, Nvidia remains the the fastest AI accelerator for every workload, on a single chip basis. While Qualcomm increased their model coverage, Nvidia was the only firm to submit benchmark results for every AI model. But the star of the Nvidia software show for inference is increasingly the Triton Inference Server, which manages the run-time optimizations and workload balancing using Kubernetes.

Source: Forbes September 22, 2021 16:52 UTC