Since R is an in-memory language, it sometimes has a reputation of not being able to handle big data. However, using some creativity and strategic thinking, you can use big data in your predictive analytics projects quite successfully.
In addition to the preceding approaches, there are currently a number of alternative approaches you may wish to research, such as:
There are packages available that avoid storing data in memory. Instead, objects are stored on hard disk and analyzed in chunks. As a side effect, the chunking also leads naturally to parallelization, if the algorithms allow parallel analysis of the chunks in principle. You can search: Revolution R Enterprise for some background on the topic.
Integrating higher performing programming languages is becoming a popular alternative to dealing with big data sources in R. This concept takes portions of R code and moves them to another language that may be better suited to carry out the logic or work than R is. This blends the best of R while avoiding performance bottlenecks.
This outsourcing of code chunks from R to another language can easily be hidden in functions. In this case, proficiency in other programming languages is mandatory for the developers, but not for the users of these functions.