try this... something like this should split your fasta file (called bigfasta here) into several small fasta files based on the sequence display id (i assume your sequences have nice looking identifiers). learn to use bioperl to your advantage for reading and writing biological sequence files... its been done by the excellent bioperl project, and you don't want to reinvent the wheel everytime.

use strict;    # always...
use Bio::SeqIO;

my $bigfasta = "bigfasta.faa";
my $seqin = Bio::SeqIO->new(-file => $bigfasta, -format=>"fasta");

while ($inseq = $seqin->next_seq) {
        my $id = $inseq->display_id;
        my $outfile = "$id.fasta";
        my $seqout = Bio::SeqIO->new(-file=>">$outfile", -format=>"fas
+ta");
        $seqout->write_seq($inseq);

}
[download]