Hi 

I have 2500 CEL files, I put them in my work directory and then I ran `ReadAffy`()`, but It gave me "cannot allocate vector of size of... error , I've tried with 500 CEL file and I gave the same error again,please let me know about your experience in this case
Thanks in advance

You don't say what kind of arrays, nor do you say if they are related enough that you should rationally process them all together. It's always possible that somebody ran that number of arrays for a single study, in which case you would want to process them in a single batch, but if you are combining a bunch of data from disparate sources, it is not likely to be a good idea.

That said, there are ways to process large numbers of files. There is the frma package, which you can use to process subsets that are small enough that your computer can process them, which could be pretty simple if you are using arrays for which there is already a frmavec package (see here and search for frmavec).

There is also aroma.light, which you can use to normalize and summarize pretty much any number of arrays, contingent upon having enough disk space for all the files you will be creating.

Another alternative would be to spin up a monster AMI with enough RAM to process all the arrays in one go. The AMI might be the more reasonable way to go - it wouldn't really cost that much, and once you have the ExpressionSet you are probably going to be fine with your desktop computer, unless it's really RAM deficient.