Performance Tweaks: Difference between revisions
Kotakomputer (talk | contribs) No edit summary |
Kotakomputer (talk | contribs) m (→VZdump) |
||
Line 74: | Line 74: | ||
=VZdump= | =VZdump= | ||
By default vzdump limited to 10000 Kbps (10 Mbps). That's why some | By default vzdump limited to 10000 Kbps (10 Mbps). That's why some users confuse with the slower data transfer on Proxmox. To increase the VZdump speed you need to edit: | ||
# nano /etc/vzdump.conf | # nano /etc/vzdump.conf | ||
Find out: | Find out: | ||
Line 80: | Line 80: | ||
bwlimit: 10000 | bwlimit: 10000 | ||
Change to any value you like | Change to any value you like, to change bandwidth limit to 50 Mbps: | ||
bwlimit: 50000 |
Revision as of 12:45, 29 September 2012
Note: Article about Proxmox VE 2.0 |
Introduction
This page is intended to be a collection of various performance tips/tweaks to help you get the most from your virtual servers.
KVM
USB Tablet Device
Disabling the USB tablet device in windows VMs can reduce idle CPU usage and reduce context switches. [1]
VirtIO
Use virtIO for disk and network for best performance.
Linux has the drivers built in
Windows requires drivers which can be obtained here: Fedora VirtIO Drivers
Disk Cache
Note: The information below is based on using raw volumes, other volume formats may behave differently. |
cache=none seems to be the best performance and is the default in Proxmox 2.X.
host don't do cache. guest disk cache is writeback Warn : like writeback, you can loose datas in case of a powerfailure you need to use barrier option in your linux guest fstab if kernel < 2.6.37 to avoid fs corruption in case of powerfailure.
cache=writethrough
host do read cache guest disk cache mode is writethrough Writethrough make a fsync for each write. So it's the more secure cache mode, you can't loose data. It's also the slower.
cache=directsync.
host don't do cache. guest disk cache mode is writethrough similar to writethrough, a fsync is made for each write.
cache=writeback
host do read/write cache guest disk cache mode is writeback Warn : you can loose datas in case of a powerfailure you need to use barrier option in your linux guest fstab if kernel < 2.6.37 to avoid fs corruption in case of powerfailure.
caching adds additional data copies and bus traffic causing it to perform worse.
For read cache memory: try to add more memory in your guest, they already do the job with their buffer cache
cache=directsync and writethrough can be fast, if you have a san or raid controller with cache with battery by exemple
Avoid to use cache=directsync and writethrough with qcow2 files.
some interestings articles :
barriers : http://monolight.cc/2011/06/barriers-caches-filesystems/
cache mode and fsync : http://www.ilsistemista.net/index.php/virtualization/23-kvm-storage-performance-and-cache-settings-on-red-hat-enterprise-linux-62.html?start=2
Application Specific Tweaks
Microsoft SQL Server
Trace Flag T8038
Setting the trace flag -T8038 will drastically reduce the number of context switches when running SQL 2005 or 2008.
To change the trace flag:
- Open the SQL server Configuration Manager
- Open the properties for the SQL service typically named MSSQLSERVER
- Go to the advanced tab
- Append ;-T8038 to the end of the startup parameters option
For additional references see: Proxmox forum
VZdump
By default vzdump limited to 10000 Kbps (10 Mbps). That's why some users confuse with the slower data transfer on Proxmox. To increase the VZdump speed you need to edit:
- nano /etc/vzdump.conf
Find out:
- bwlimit: KBPS
bwlimit: 10000
Change to any value you like, to change bandwidth limit to 50 Mbps: bwlimit: 50000