What do you want to do with it once you have read it in? There is no
point in reading in a matrix that is likely to take up all of the
available memory (160Mb if numeric) ....
Possible suggestions are to load it into a database and access in blocks
of rows or columns, or spend a small amount of money on another 1Gb of
memory and write C code in a package to read the file.
On Sat, 1 Feb 2003 TyagiAnupam at aol.com wrote:
> I am looking for suggestions to read a large fixed-column format file and
> create a data-frame with about 50,000 observations and 400 variables, each
> variable in 2--5 colums. This is on a machine with 128 Mb of RAM with R on
> Linux or MS-win. I will also be thankful for any other suggestions about
> working with data of this size.
--
Brian D. Ripley, ripley at stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595