Package 'ff' supports fast on-disk objects: standard and packed atomics, ffdf data.frames, csv import/export, chunking and indexing. Integrated package 'bit' supports fast in-memory booleans: 'bit' (1-bit boolean), 'bitwhich' (asymetric booleans).
The project summary page you can find here.
presentations on available functionality
Oehlschlägel (2010) Managing large datasets in R - ff examples and concepts. Invited presentation at Institute for Statistics and Mathematics, Vienna University of Economics and Business. Example source code
Oehlschlägel, Adler (2009) Coordinating package 'ff' for large objects with R base. Presentation at The sixth international workshop on Directions in Statistical Computing (DSC 2009), Copenhagen
Oehlschlägel, Adler (2009) Managing data.frames with package 'ff' and fast filtering with package 'bit'. Presentation at UseR!2009, Agrocampus, Rennes
Adler, Oehlschlägel, Nenadic, Zucchini (2008) Large atomic data in R package 'ff'. Presentation at UseR!2008, statistics department, University of Dortmund
Adler, Oehlschlägel, Nenadic, Zucchini (2008) High-Performance Processing of Large Data Sets via Memory Mapping: A Case Study in R And C++. Presentation at Joint Statistical Meetings, Denver, Colorado.
presentations on future functionality
Oehlschlägel, Adler, Nenadic, Zucchini (2008) A first glimpse into 'R.ff'. Presentation at UseR!2008, statistics department, University of Dortmund