It's an oldie from IBM but still as good as ever!
download nmon from ibm and their unsupported analyzer spreadsheet.
run
./nmon -f -s2 -c 60
where c is number of datapoints
and s is interval between gatherpoints
f is for background.
then sort the file into a .csv
open nmon analyser v33C.xls
import system.090914_1445.csv into it.
Aswell did a little perl-script for merging logs if you've as me managed to collect to much data for excel to handle.
Just change $a % 50 to a value that fits your needs
#!/usr/bin/perl
open(FILE, shift) or
die("Unable to open file!");
$a = 0;
$b = 0001;
$start = 1;
foreach $sak (<FILE>) {
chomp($sak);
@divided = split(/,/,$sak);
$curr = $divided[0];
if (($curr =~ "AAA") || ($curr =~ "BBBP")) {
print $sak . "\n";
next;
}
if ($start == 1) {
$last = $curr;
$start = 0;
}
$a++;
if ($curr !~ $last) {
$a = 0;
$b = 0001;
$last = $curr;
print $sak . "\n";
next;
}
if ($a % 50 == 0) {
$d = sprintf("%04d",$b);
$divided[1] = "T$d";
print join(',', @divided) . "\n";
$b++;
}
}
close(FILE);
