February 17, 201412 yr For those of you wishing to find out where Xen is headed, RC4.4 just got released. We run 4.3 in unRAID 6.0 currently... http://www.phoronix.com/scan.php?page=news_item&px=MTYwNjg
February 17, 201412 yr It's not a Release Candidate / Beta... Its an Official Release. Xen 4.4 went into feature freeze in October and since then they did the beta testing, release candidates, bug fixes, etc. Speaking of Xen... We are running Xen 4.3.1 and Xen 4.3.2 was released Feb 14. Below is a list of all the changes / bug fixes in 4.3.2 http://xenbits.xen.org/gitweb/?p=xen.git;a=shortlog;h=refs/tags/RELEASE-4.3.2
February 17, 201412 yr Unless the RC wiki is wrong I think Ironic is dead on. See here last updated today. http://wiki.xen.org/wiki/Xen_4.4_RC4_test_instructions. Looks like they have a little way to go to fix the bugs. burt
February 18, 201412 yr Unless the RC wiki is wrong I think Ironic is dead on. See here last updated today. http://wiki.xen.org/wiki/Xen_4.4_RC4_test_instructions. Looks like they have a little way to go to fix the bugs. burt My bad... The Release will be out by the end of the month but still working through some issues and the latest Release Candidates is version 4. Tom has to wait for someone to create a build package for it or figure it out himself (no easy task!).
February 18, 201412 yr Author You just concentrate on driving that bus before you hurt yourself. Sent from my Nexus 5 using Tapatalk
February 18, 201412 yr Can anyone translate to English the reason 4.4 is gonna be so much better? I see a lot of writing and I would have to do more reading than it is worth to decrypt it all. Thanks
February 18, 201412 yr Can anyone translate to English the reason 4.4 is gonna be so much better? I see a lot of writing and I would have to do more reading than it is worth to decrypt it all. Thanks http://www.phoronix.com/scan.php?page=news_item&px=MTU3OTg In few words, it will allow OSes that doesn't support full PV to run faster with less overhead. It means Windows or BSD guests will have better performance, using paravirtualized NICs, USB etc.
February 18, 201412 yr Ok, sweet. And if they are hvm does that mean the benefit is not so much HVM guests are the target. You will have HVM guests with PV I/O, not emulated through QEMU. E.g., uning nics in HVM guests will consume much less CPU resources than it does today.
February 18, 201412 yr Any word on Nvidia GPU pass thru in RC 4.4? Sent from my LG-VS980 using Tapatalk
February 18, 201412 yr quite excited by PVH support, sounds like potential for performance boost, watched this presentation, way over my head on some of this but still very interesting. http://www.xenproject.org/component/allvideoshare/video/latest/xensummit-pvh-future-of-pv.html quote from it, "out of the three methods PVH should be the fastest, the three methods being HVM with PV drivers, pure PV, and then PVH" so interesting to see what it will be like.
March 9, 201412 yr Ok, sweet. And if they are hvm does that mean the benefit is not so much HVM guests are the target. You will have HVM guests with PV I/O, not emulated through QEMU. E.g., uning nics in HVM guests will consume much less CPU resources than it does today. I don't read it that way, seems to me fully PV guests are the target. http://wiki.xen.org/wiki/Xen_Overview PV in an HVM Container (PVH) - New in Xen 4.4 Xen 4.4 will bring a virtualization mode called PVH. This is essentially a PV guest using PV drivers for boot and I/O. Otherwise it uses HW virtualization extensions, without the need for emulation. First patches are going into xen-unstable just after the Xen 4.3 release and functionality should be at least available as a pre-view in Xen 4.4. PVH has the potential to combine the best trade-offs of all virtualization modes, while simplifying the Xen architecture. This also has a useful diagram: http://wiki.xen.org/wiki/Virtualization_Spectrum So for a Windows guest this new mode doesn't seem to offer anything, it will still run in HVM mode and they you can use PV drivers for increased performance as now. But confusingly the description of PVHVM mode in the above document contradicts the mode as described here http://docs.oracle.com/cd/E27300_01/E27309/html/vmusg-vm-modes.html , so who knows?!
Archived
This topic is now archived and is closed to further replies.