In this completely demo-based talk, Chris Fregly from PipelineIO will demo the latest 100% open source research in high-scale, fault-tolerant, distributed model training, testing, and serving using Tensorflow, Spark ML, Jupyter Notebook, Docker, Kubernetes, and NetflixOSS Microservices. This talk will discuss the trade-offs of mutable vs. immutable model deployments, on-the-fly JVM byte-code generation, global request batching, miroservice circuit breakers, and dynamic cluster scaling - all from within a Jupyter notebook. All code and docker images are available from Github and DockerHub at http://pipeline.io.
