Hello everybody!
We have a problem with a linux server that crashes when we try to read large
datasets in R.
The R code is as followed:
komplett <- read.spss("komplett2003aar.sav", to.data.frame =TRUE,
reencode ="Latin1")
The information about the linux server is:
Linux version 2.6.24-19-generic (buildd at king) (gcc version 4.2.3
(Ubuntu 4.2.3-2 ubuntu7)
We had a memory of 15.7 GiB and swap of 5.6 GiB when the server crashes. We used
about 10 GiB of the memory and the all the swap when we read in the dataset.
We have now expand the swap to 25.6 GiB, but with the same memory. With these
changes are we now able to read the datasets. We are not sure if that's the
reason, but hope so!! Have any one experience the same? Can the size of the swap
have that influence on large dataset. The dataset we tried to read was 837.2 MB.
Thanks!!!
Regards,
Katrine
