Dear Jatin, Maybe it’s a good idea first to implement Spice: <video> <model type='qxl' ram='65536' vram='65536' heads='1'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/> </video> <channel type='spicevmc'> <target type='virtio' name='com.redhat.spice.0'/> <address type='virtio-serial' controller='0' bus='0' port='1'/> </channel> Spice should be installed on the host. Do you use virtio ? Greetings, Dominique. Van: Jatin Davey [mailto:jashokda@cisco.com] Verzonden: dinsdag 14 april 2015 12:23 Aan: libvirt-users(a)redhat.com Onderwerp: [libvirt-users] VM Performance using KVM Vs. VMware ESXi Hi All We are currently testing our product using KVM as the hypervisor. We are not using KVM as a bare-metal hypervisor. We use it on top of a RHEL installation. So basically RHEL acts as our host and using KVM we deploy guests on this system. We have all along tested and shipped our application image for VMware ESXi installations , So this it the first time we are trying our application image on a KVM hypervisor. On this front i have done some tests to find out how our application's response time is when deployed on KVM and then compare it with a VM deployed on VMware ESXi. We have a benchmark test that basically loads the application simulating a load of 100 parallel users logging into the system and downloading reports. These tests basically use a HTTP GET query to load the application VM. In addition to that i have taken care to use the same hardware for both the tests , one with RHEL(Host)+KVM and another with VMware ESXi. All the hardware specifications for both the servers remain the same. The load test also remains the same for testing with both the servers. First observation is that the average response time on the VMware ESXi is : 500 milli-seconds while the application's average response time when deployed using RHEL(Host)+ KVM is : 1050 milli-seconds. The response time of the application when deployed on KVM is twice as much as when it is deployed using VMware ESXi. I did few more tests to find which sub-system on these servers shows varying metrics. First i started with IOZone to find out if there is any mismatch in the speed with which data is read / written to the local disk on the two VMs and found that "Read" speed in the VM that was deployed using RHEL(Host)+KVM was twice as slow as the VM which was deployed using VMware ESXi. For more on IoZone , Please refer : http://www.iozone.org/ more specifically the following IoZone metrics were twice as less when compared to the server running with VMware ESXi: Read Re-read Reverse-Read Stride Read Pread Note: I had run the IoZone tests on the VMs on both the servers. Second observation to be made was the output from the "top" command. I could see that the VM deployed on RHEL(Host)+KVM was showing high numbers for the following metrics when compared with the VM deployed on VMware ESXi: load averages %sy for all the logical processors %si for all the logical processors i debugged further to find out which device is causing more interrupts and found it to be "ide0" , See the output from the /proc/interrupts file below: The other interrupts apart from ide0 are pretty much similar to the VM deployed using VMware ESXi. ************/proc/interrupts ******************* [root@localhost ~]# cat /proc/interrupts CPU0 CPU1 CPU2 CPU3 CPU4 CPU5 CPU6 CPU7 0: 795827 0 0 0 0 0 0 0 IO-APIC-edge timer 1: 65 0 0 0 0 0 0 0 IO-APIC-edge i8042 6: 2 0 0 0 0 0 0 0 IO-APIC-edge floppy 8: 0 0 0 0 0 0 0 0 IO-APIC-edge rtc 9: 0 0 0 0 0 0 0 0 IO-APIC-level acpi 10: 425785 0 0 0 0 0 0 0 IO-APIC-level virtio0, eth0 11: 47 0 0 0 0 0 0 0 IO-APIC-level uhci_hcd:usb1, HDA Intel 12: 730 0 0 0 0 0 0 0 IO-APIC-edge i8042 14: 188086 0 0 0 0 0 0 0 IO-APIC-edge ide0 NMI: 0 0 0 0 0 0 0 0 LOC: 795813 795798 795783 795767 795752 795737 795723 795709 ERR: 0 MIS: 0 ********************************************* Any pointers to improving the response time for the VM for RHEL(Host)+KVM installation would be greatly appreciated. Thanks Jatin

About Spice: I think it’s good practice to use spice because it improves the performance of the VM in general by improving screen performance. If your VM is constantly displaying output, you’ll probably will notice a difference. About virtio: You can see it in the settings. Better yet, it’s in your XML. If you post your XML, we can take a look… Van: Jatin Davey [mailto:jashokda@cisco.com] Verzonden: dinsdag 14 april 2015 12:44 Aan: Dominique Ramaekers; libvirt-users(a)redhat.com Onderwerp: Re: [libvirt-users] VM Performance using KVM Vs. VMware ESXi On 4/14/2015 4:02 PM, Dominique Ramaekers wrote: Dear Jatin, Maybe it’s a good idea first to implement Spice: <video> <model type='qxl' ram='65536' vram='65536' heads='1'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/> </video> <channel type='spicevmc'> <target type='virtio' name='com.redhat.spice.0'/> <address type='virtio-serial' controller='0' bus='0' port='1'/> </channel> Spice should be installed on the host. [Jatin] How will using Spice help in improving better response time from a web application that i am using inside a VM ? Can you throw more light in this regard if it helps in improving the response time. Do you use virtio ? [Jatin] I am not sure about this. Can you tell me how can i find out if i am using virtio or not ? Greetings, Dominique. Thanks Jatin

On Tue, Apr 14, 2015 at 04:53:52PM +0530, Jatin Davey wrote: This disk is configured to use IDE, so performance of anything that does disk I/O is going to be terrible. You really want to be using virtio. This doesn't have any model listed at all, so it will be falling back to a generic emulated NIC. Again performance of this is likely going to be terrible for anything doing network I/O. You want to be using virtio for this too. Regards, Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|

Please read: https://libvirt.org/formatdomain.html -----Oorspronkelijk bericht----- Van: Jatin Davey [mailto:jashokda@cisco.com] Verzonden: dinsdag 14 april 2015 13:39 Aan: Daniel P. Berrange CC: Dominique Ramaekers; libvirt-users(a)redhat.com Onderwerp: Re: [libvirt-users] VM Performance using KVM Vs. VMware ESXi On 4/14/2015 4:58 PM, Daniel P. Berrange wrote: How do i make use of virtio for the both disk and network that you have mentioned above ? Any pointers to it would be helpful. Thanks Jatin

On 2015-04-14 14:33, Jatin Davey wrote: Unless their kernels are very old, you only need to rebuild your initrd as documented in the wiki (or similarly for other distributions). -- Mit freundlichen Grüßen, / Best Regards, Sven Schwedas Systemadministrator TAO Beratungs- und Management GmbH | Lendplatz 45 | A - 8020 Graz Mail/XMPP: sven.schwedas(a)tao.at | +43 (0)680 301 7167 http://software.tao.at

On 4/14/2015 6:32 PM, Tom Hughes wrote: My VM is using the kernel 2.6.18-164.el5 [root@localhost ~]# uname -a Linux localhost 2.6.18-164.el5 #1 SMP Thu Sep 3 03:28:30 EDT 2009 x86_64 x86_64 x86_64 GNU/Linux Should this be fine ? Thanks Jatin

On 4/14/2015 6:51 PM, Tom Hughes wrote: [Jatin] Yes , it does not have virtio-scsi [root@localhost virtio]# modinfo virtio-scsi modinfo: could not find module virtio-scsi [Jatin] [root@localhost virtio]# modinfo virtio-blk filename: /lib/modules/2.6.18-164.el5/kernel/drivers/block/virtio_blk.ko license: GPL description: Virtio block driver alias: virtio:d00000002v* srcversion: A9DBCDB63CCE543B4BBAF31 depends: virtio vermagic: 2.6.18-164.el5 SMP mod_unload gcc-4.1 module_sig: 883f3504a9f767957f09578a977b7e1121870a0c3c17c4c17a09f4825b9b3add41f93c65bf75ad0a0afa2471b36c74a6c7dedc468dd197c1d9eea86aa Thanks Jatin

I am looking at upgrading to RHEL 7 or 6.6 but before getting there i want to see if the existing version works fine. If it works then i am good but if it does not then i will have to upgrade to newer versions. Thanks Jatin

Jatin, Using qemu without the virtio scsi and nic drivers is like running vmware with ide disks and e1000 nic instead of LSI disks and vmxnet3 nics, it forces the system to emulate completely different hardware. In linux the virtio drivers are implemented in the kernel, so you either need a new kernel or the virtio kernel modules. I'm not sure which for RHEL5, but I suspect you can get what you need with the kmod-virtio package, then rebuilding the initrd image to load the modules at boot. Keep in mind that sda will change to vda, which on some linux distros requires updates to the bootloader and inittab, but redhat uses filesystem labels if I remember correctly, so it should just work. Please let us know how the benchmarks look after you get virtio working, I'm curious... schu On 4/14/2015 5:33 AM, Jatin Davey wrote:

On 4/15/2015 2:21 AM, Nikki VonHollen wrote: Sure , Thanks Nikki. I will surely have to dig deep into storage as i can see a clear bottleneck in it. Thanks for the pointers. Regards, Jatin