A Docker Container Toolbox for the Data Scientist
summary
A major U.S. bank deploys containers that include R Python and SAS so data scientists get browser access to Jupyter RStudio and SAS Studio; a SAS runtime fits into a ~9 GB container (comparable to existing 7 GB images) and runs successfully in production at three large banks. The team ships a sas jupyter kernel and sas pi to let notebooks call SAS math, and pushes scoring and data-step jobs to a 40-node Hadoop cluster via YARN so models run in-place without moving data.
AI-written summary · about
chapters (4)
related talks
45mData Science Apps: Beyond Notebooks
46mJump-Start Your Data Science Learning Java & Jupyter Notebooks
21mA guide to modern reproducible data science with R
19mData Science in a Cloud World: What Every Data Scientist Needs to Know
42mDevelopment of Data Science Ecosystem for Ruby
42mon SageMaker
17mBuilding a data science pipeline for the FT with RStudio Connect
49mExperimenting with Large Data Volumes Using Notebooks