Hello,
currently I'm setting up my new home 24/7 server which has become an AMD Athlon 
system because of its energy efficience. Because XEN - due to its outdated 
kernel - does not work on my hardware, KVM was kind too fresh when I started my 
project and the license information of VirtualBox was not clear I decided to 
use UML.

To make it short: Everything works stable but way too slow. The effect is that 
in my opinion every larger memory allocation leads to heavy disk activity. E. g.

dd if=/dev/zero of=/dev/null bs=128M count=1

where no disk I/O is done leads to heavy write access on the hosts harddisks 
(due to its typical access pattern on the hosts RAID1 system disks I can tell 
quite for sure that it's write access). After approx. 5s dd returns and states 
very low performance of about 8..14MB/s. When I use small blocks of about 1M 
and increase the count instead performace boosts a lot.

The UML has enough memory (256M, ~190M free), the host itself has 6GB (well, I 
started with 2GB before I experimented with VirtualBox and a Windows 
installation - the problems existed from the beginning when the machine only 
had 2GB). Swap space is not used, even when I turn of all swap space on the 
host and the UML heavy disk access appears. For testing I moved the UML's file 
system to a separate disk, even then on the host's system disks heavy I/O 
happens every time I do this call.

Any idea what the reason for this behaviour could be?

With kind regards,
Torsten Lang

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
User-mode-linux-user mailing list
User-mode-linux-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-user

Reply via email to