xgrr.de – the whole not the half

things, thoughts and stuff out of life, daily business and computer science


Some VMware related I/O speed optimizations

While studying the documentation to Lustre (a distributed Linux filesystem) I found this:

Choosing an alternate scheduler is an absolutely necessary step
to optimally configure Lustre for the best performance. The “cfq” and “as” schedulers should never be used
for server platform.

So I revised my VMware installation and found out that all my guest are configured to use the deadline scheduler. But as it seemed my host did use the cfq. In the user guide there was described how to change the used scheduler while running the system (where hda is the disk):

[root@cfs2]# cat /sys/block/hda/queue/scheduler
noop [anticipatory] deadline cfq
[root@cfs2 ~]# echo deadline > /sys/block/hda/queue/scheduler
[root@cfs2 ~]# cat /sys/block/hda/queue/scheduler
noop anticipatory [deadline] cfq

So I changed the scheduler and it seems the overall performance is dramatically increased.

noatime: Rocks like hell

For my new VMware environment I wanted to make the harddisks faster because it is the only thing which slows down the whole system. For the running virtual machines I enabled the “noatime” function for the ext3 partitions and put the ext3 journal in writeback mode which speeds up things even more.

Important was the noatime function which stores the time every time a file is accessed. For a webserver with a gallery or a lot static files this is a huge performance hit of course.

Tuning up VMware Server on Linux

When I finished my VMware Server 1.0.4 installation on Debian Etch with a 2.6.22.9 Kernel and moved all customers to the new server I was very disappointed. The overall performance was very poor on a dual Opetorn system with 2 GB memory and I had no explanation for it. After a little testing I figured that I gave all the available memory to use for the VMs and left almost nothing over for the server itself. When I pushed about 100 MB of memory free for the server itself the performace improved a lot.

But after all. A VMware Server with six running VMs doesn’t fit into 2 GB of memory. Start with three, be certain with four GB of main memory.

Fast, cool and shiny: ADSL2+

Jippie. Nie mehr lange auf irgendwas warten. Donnerstag habe ich zu Hause endlich richtig schönes breitband Internet bekommen. Mit bis zu 2,0 Mbyte pro Sekunde Downloaden ist eine Wohltat.