Hi all, On Thu, 02 May 2013 18:34:10 +0200 (CEST) Alexandre DERUMIER <[email protected]> wrote:
>
> Maybe can we compute average source new writes (ios ? bandwith ?), or
> something like that, and decrease progressively the ios or bandwith.
> I need to think about that a little more.
>
I have given this some thoughts and I think it could be implemented
this way:
my $max_bps_wr = get_target_max_bps_wr($target);
my $bps_wr = $max_bps_wr;
my $new_bps_wr = 0;
my $timer = 0;
eval {
while (1) {
my $elapsed = time() - $timer
if ($elapsed < $timer) {
$new_bps_wr = $bps_wr * 0.9;
if ($new_bps_wr <= ($max_bps_wr * 0.1) {
die "Target $target is to congested";
}
}
else {
$new_bps_wr = $bps_wr * 1.05;
if ($new_bps_wr > $max_bps_wr) {
$new_bps_wr = $max_bps_wr;
}
}
$timer = $elapsed;
$bps_wr = $new_bps_wr;
PVE::QemuServer::qemu_block_set_io_throttle(
$vmid, $target, 0, 0, $new_bps_wr);
# do other migration stuff
}
};
do {
my $err = $@;
PVE::QemuServer::qemu_block_set_io_throttle(
$vmid, $target, 0, 0, 0);
# do other error stuff
die $err;
} if $@;
PVE::QemuServer::qemu_block_set_io_throttle($vmid, $target, 0, 0, 0);
--
Hilsen/Regards
Michael Rasmussen
Get my public GnuPG keys:
michael <at> rasmussen <dot> cc
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E
mir <at> datanom <dot> net
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C
mir <at> miras <dot> org
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917
--------------------------------------------------------------
People think love is an emotion. Love is good sense.
-- Ken Kesey
signature.asc
Description: PGP signature
_______________________________________________ pve-devel mailing list [email protected] http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
