Choosing the Right Infrastructure for Production Machine Learning скачать в хорошем качестве

Choosing the Right Infrastructure for Production Machine Learning

Finding the right combination of infrastructure to support production AI at-scale can be time consuming and costly. In addition to identifying what kind of hardware can best support your production needs, picking the right deployment paradigm can save you thousands in cloud compute costs. This tech talk walks you through evaluating the infrastructure requirements of ML workloads to identify the right combination to support your team’s needs for running inferences in production, at-scale. For more resources, visit https://www.modzy.com/modzy-blog. #machinelearning #artificialintelligence #ml #infrastructure #edgeai #edgecomputing #cpu #tpu #fpga #gpu 0:00 Introduction 3:03 Questions to Ask Yourself 5:10 Von Neumann Architecture 9:17 Instructions and Architectures 14:59 GPUs As Accessories 18:40 Inferentia Chips 22:22 Performance Comparisons 23:47 Choosing the Right Hardware 27:07 Factoring it All Together.. 30:30 Recommended Reading