Third as suggested Compress::Zlib could be an option (in general fast and good compression). But there are many others on CPAN. In general it depends on what you try to compress (structure of the string), and the size of it. You don't give info on those.

Thanks your reply.
1)The big string ,i mean the string is large enough,about 8-10k
2)I use the SqlServer ( varchar(max) ) to store the string.
3)I just compress common string ,eg.:

my $string = "
aaaaaaaaaaaaaaaaaaaaaaaa  sdfsdf
bbbbbbbbbbb sdfsdfsdf
sdfffffffff                   sdfsdf < tabl> 
sfsdfffffffffffffffffffffffffffffffffffffffffff
...
much and more ....
"
[download]

1)The big string ,i mean the string is large enough,about 8-10k

This doesn’t strike me as "big".

2)I use the SqlServer ( varchar(max) ) to store the string.

SQLServer should have not trouble handling those strings.

3)I just compress common string ,eg.:...

Any of the suggested compression alternatives should work for you. Based on the example you provide I assume that you will obtain excellent compression ratios.

I would like to draw your attention to one more alternative: RLE which stands for Run Length Encoding. It doesn’t give you a great compression ration but it’s very simple and therefore fast. I have found a Perl implementation for it which can be found here (I did not tested it though)

perldoc Compress::Zlib

perldoc Archive::Zip

#!/usr/bin/perl
use warnings;
use strict;
use Math::BaseCnv;
#fast

my $time = time;
print "time -> $time\n";          

my $time_128 = cnv( $time, 10, 128 );
print "time_128 -> $time_128\n";
#this will sometimes print hidden newlines
exit;
[download]