R Programming in Data Science: High Velocity Data

Course details

High-velocity data—such as the information that springs from Twitter and IoT devices—comes barreling in at a speed beyond normal comprehension, demanding high-performance from both hardware and software. While it might not initially appear up to the challenge, the R programming language can be revved up to operate with high-velocity data. Written close to the metal by sitting directly on top of the C programming language, R provides a rich set of data structures and concepts. This course drills down into efficient R programming, providing practical strategies that can help you work your mojo on high-velocity data.
Instructor Mark Niemann-Ross begins by sharing a framework for understanding the different types of high-velocity data. He then covers how to use R to acquire high-velocity data, as well as how to leverage profiling tools and optimize R code for use with high-velocity data. He wraps up by exploring how to use R to present data, including how to use Shiny—an R package that allows you to build web apps straight from R—for interactive dashboards.

Learning objectives
Explore concepts of batch data.
Identify libraries for handling near real-time data.
Recall the fundamentals of polling data.
Recognize the best approach to optimizing code.
Identify ways to avoid copying data.
Explore Flexdashboard and related tools for creating static reports.

Concepts

0. Introduction

01 - How can you use R with high-velocity data

1. Problems and Opportunities with High-Velocity Data

02 - Perspectives on high-velocity data
03 - Simulating high-velocity data
04 - Concepts of batch data
05 - Handling batch data with R
06 - Working with near real-time data
07 - Handling near real-time data with R
08 - Concepts of real-time data
09 - Handling real-time data with R
10 - Setting a default CRAN mirror

2. Using R to Acquire High-Velocity Data

11 - Polling for data in R
12 - Interrupt-driven data acquisition with R

3. Profiling Tools for R

13 - Tools
14 - Profvis
15 - Rprof
16 - microbenchmark

4. Optimizing R to Process High-Velocity Data

17 - Improving the speed of loops
18 - Optimizing if then else with ifelse
19 - Avoid copying data
20 - Combining optimizations
21 - Use RCPP to speed up functions
22 - Using microbenchmark to check results

5. Using R to Present High-Velocity Data

23 - Static and dynamic reports
24 - Use R Markdown for static dashboards
25 - Flexdashboard and other enhancements for static reports
26 - Use Shiny for interactive dashboards
27 - Use plumber to create APIs
28 - Cran task view for WebTechnologies

Conclusion

29 - Summary

R Programming in Data Science: High Velocity Data

Authors

Mark Niemann-Ross