redeeman: michaellarabel: your recent kvm benchmark article kinda sucks :) first of all, you give NO mention of what sort of disk setup you've done with KVM, which is ALL the difference, and secondly, your closing remark: "The disk benchmarks took a large hit when running Ubuntu 9.10 "Karmic Koala" within a virtual machine using KVM, but that is to no surprise considering the setup." - that makes no sense? considering WHAT setup?
bkero: would have preferred a mention to what disk driver was being used. virtio?
schestowitz: michaellarabel: I need your help.
michaellarabel: schestowitz: What is your problem>
schestowitz: You know the "bloated" remark?
michaellarabel: Which bloated remark? Or just in general terms
schestowitz: I'm looking past:
schestowitz: Linux 2.6.28 Kernel Benchmarks
schestowitz: http://www.phoronix.com/scan.php?page=article&item=linux_2628_benchmarks&num=1
schestowitz: Linux 2.6.26 Kernel Benchmarks
schestowitz: http://www.phoronix.com/scan.php?page=article&item=linux_2626_benchmarks&num=1
schestowitz: Linux 2.6.23 Kernel Benchmarks
schestowitz: http://www.phoronix.com/scan.php?page=article&item=872&num=1
schestowitz: Linux 2.6.30 Kernel Benchmarks
schestowitz: http://www.phoronix.com/scan.php?page=article&item=linux_2630&num=1
schestowitz: Linus calls Linux 'bloated and huge'
schestowitz: During a roundtable discussion at LinuxCon in Portland, Oregon this afternoon, moderator and Novell distinguished engineer James Bottomley asked Tovalds whether Linux kernel features were being released too fast, before the kernel is stabilized.
schestowitz: Citing an internal Intel study that tracked kernel releases, Bottomley said Linux performance had dropped about two per centage points at every release, for a cumulative drop of about 12 per cent over the last ten releases. "Is this a problem?" he asked.
schestowitz: http://www.channelregister.co.uk/2009/09/22/linus_torvalds_linux_bloated_huge/
schestowitz: Does your data agree with this?
redeeman: michaellarabel: what disk driver and disk format did you use for the kvm benchmark?
michaellarabel: schestowitz: It depends upon the area. Some areas of the kernel have slowed down while others have improved.
michaellarabel: redeeman: The default Ubuntu options when using virt-manager... Can confirm details shortly, but I just got back into the US last night so busy catching up with other work at the moment.
redeeman: i have no idea what virt-manager does, but you REALLY shouldn't be doing such things, if it uses qcow2 and IDE controller it will have horrible results
redeeman: and it does in NO way represent actual performance
Ivanovic: hi michaellarabel
michaellarabel: Hi ivanovic
Ivanovic: so have you enjoyed octoberfest?
michaellarabel: Ivanovic: Yes, Oktoberfest was excellent.