How Cookpad Leverages Triton Inference Server to Boost Model Serving

Discover how Cookpad optimizes its machine learning model deployment using Triton Inference Server in this 32-minute conference talk. Learn about the challenges faced by Machine Learning Platform teams when scaling model deployment, including managing diverse frameworks and infrastructure requirements. Explore how Triton Inference Server, an open-source tool from Nvidia, simplifies the deployment process and improves resource utilization. Gain insights into deploying concurrent models on single GPU or CPU and multi-GPU servers, and understand how Cookpad's ML Platform Engineers leverage this technology to boost their model serving capabilities.