R and Big Data (Databases)

Presentation for Big Data London, 2019 on using big data in R via database connections.

Transcript

1. You Love Big Data! So Does R. https://github.com/akgold/bdl_2019 @alexkgold Alex K Gold #rstats
2. R For Big Data Single-Threaded Just Slow TRUE In-Memory DON’T MATTER @alexkgold
3. Single-Threaded Just Slow In-Memory • doFuture • RStudio Server Pro Launcher • Fast Enough (profvis) • Rcpp @alexkgold
4. In-Memory Database Dev Machine To Scale @alexkgold
5. Big Data Strategies for R @alexkgold
6. Strategy 1: Sample and Model Use favorite R modeling package (Caret/Parsnip/rsample). Really good for iterating/prototyping. ☹ Requires care for sampling and scaling. ☹ Not good for BI tasks. @alexkgold
7. Data: London Planning Permissions 2006-current Demo! @alexkgold Model
8. Strategy 2: Chunk and Pull Great when discrete chunks exist. Facilitates parallelization. ☹ Can’t have interactions between chunks. ☹ Eventually pull in all data. . . . @alexkgold
9. Demo! @alexkgold
10. Strategy 3: Push Compute to Data Take advantage of database strengths. Get whole dataset, but move less data. ☹ Operations might not be permitted in database. ☹ Maybe your database is slow? @alexkgold
11. Demo! @alexkgold
12. 3 Big Data Strategies for R 1. Sample and Model 2. Chunk and Pull 3. Push Compute to Data But why R? ➔ None are perfect! @alexkgold
13. Import Tidy Visualize Transform Model Communicate or Automate Discovery Cycle @alexkgold
14. What about deployment? Open-Source (Free!) • Build-your-own • Shiny Server Enterprise Products • RStudio Connect • Evals @alexkgold
15. Recommendation Summary Problem Solution Single-Threading • Many R packages ◦ My favorite: doFuture • RStudio Server Pro Job Launcher R is Slow • Profile with profvis • Write in a faster language, call from R (Rcpp) In-Memory Data • Adopt a big data paradigm for R 1. Sample and Model 2. Chunk and Pull 3. Push Compute to Data @alexkgold https://github.com/akgold/bdl_2019 db.rstudio.com spark.rstudio.com therinspark.com