Re: [Nbd] nbd for large scale diskless linux -> tests & problem
- To: Wouter Verhelst <wouter@...3...>
- Cc: nbd-general@lists.sourceforge.net
- Subject: Re: [Nbd] nbd for large scale diskless linux -> tests & problem
- From: dsuchod <dsuchod@...65...>
- Date: Sat, 8 Oct 2005 22:45:24 +0200 (CEST)
- Message-id: <Pine.LNX.4.61.0510082232210.4864@...67...>
- In-reply-to: <20051008064918.GF4157@...39...>
- References: <Pine.LNX.4.61.0510072304480.8781@...66...> <20051008064918.GF4157@...39...>
Hi!! I did some more tests:
Booting of 15 machines simultaniously (each connected via separate 100Mbit
switch port, server on gigabit uplink of that switch) got me the
following:
Booting rather fast, 231MByte down 6MByte up for each client compared to
nfs with 312MByte down and 13Mbyte up. Then on six clients: cat /dev/nb0>
/dev/null (little bit pervert :-))
root 4538 2.4 0.0 3588 680 ? S 22:21 0:16
/bin/nbd-server 5
root 4627 4.3 0.0 3588 680 ? S 22:23 0:23
/bin/nbd-server 5
root 4693 3.9 0.0 3588 680 ? S 22:23 0:20
/bin/nbd-server 5
root 4696 3.3 0.0 3588 680 ? S 22:23 0:17
/bin/nbd-server 5
root 4698 0.1 0.0 3588 680 ? S 22:23 0:00
/bin/nbd-server 5
root 4699 0.1 0.0 3588 680 ? D 22:23 0:00
/bin/nbd-server 5
root 4700 3.4 0.0 3588 680 ? S 22:24 0:17
/bin/nbd-server 5
root 4701 0.1 0.0 3588 680 ? S 22:24 0:00
/bin/nbd-server 5
root 4708 0.1 0.0 3588 680 ? S 22:24 0:00
/bin/nbd-server 5
root 4709 0.1 0.0 3588 680 ? S 22:24 0:00
/bin/nbd-server 5
root 4710 4.0 0.0 3588 680 ? D 22:24 0:19
/bin/nbd-server 5
root 4762 0.3 0.0 3588 680 ? S 22:26 0:01
/bin/nbd-server 5
root 4763 0.3 0.0 3588 680 ? S 22:26 0:01
/bin/nbd-server 5
root 4764 0.3 0.0 3588 680 ? S 22:26 0:01
/bin/nbd-server 5
So no much cpu consumption, but some of them waiting for IO. During bootup
the servers gigabit link got up to 360MBits, little bit less with the
"cat" ... With all the "cats" after a while around 230MBits ... A client
is still operable (graphical login into KDE) while cat is running. I
have to purchase faster disks/and server memory now ...
I think nbd could be a real killer for nfs :-)
But I had a problem within my ramdisk setting up the nbd on some clients
(dell 2,8Ghz P4, 1GB Ram, broadcom5700 gbit ethernet). I had to add a
"sleep 10" after powering up nbd-client server 5000 /dev/nb0 ...
Otherwise the system would crash because the device was not ready for
ext2 mount. It seems that the autoconfiguratin of the ethernet link
interupts the traffic and the nbd-client can not cope with that :-(
But this rather short delay is not a big issue. Machines with other
network devices need only a sleep of 0 to 4 seconds ...
Have a nice sunday!!
Dirk
Reply to: