yue
2011-Apr-23 07:47 UTC
[Xen-devel] if hvm better than pv?why dd in hvm vm no larger than 50M, while pv greater than 50
1.if passthrough pci is default in hvm? pv disk throughput better than hvm,using dd test thanks _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Konrad Rzeszutek Wilk
2011-Apr-25 13:36 UTC
Re: [Xen-devel] if hvm better than pv?why dd in hvm vm no larger than 50M, while pv greater than 50
On Sat, Apr 23, 2011 at 03:47:23PM +0800, yue wrote:> 1.if passthrough pci is default in hvm? pv disk throughput better than hvm,using dd test > thanksPlease write your question more clearly. I am having a hard time understanding what it is that you are asking. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
yue
2011-Apr-26 08:48 UTC
Re:Re: [Xen-devel] if hvm better than pv?why dd in hvm vm no larger than 50M, while pv greater than 50
1.i find pv disk'throughput is better than hvm. i want to know why? 2.pci passthrough is said to improve performance, if it is default method of IO-process. 3.do you familar with ocfs2's source code? thanks At 2011-04-25 21:36:04，"Konrad Rzeszutek Wilk" <konrad.wilk@oracle.com> wrote:>On Sat, Apr 23, 2011 at 03:47:23PM +0800, yue wrote: >> 1.if passthrough pci is default in hvm? pv disk throughput better than hvm,using dd test >> thanks > >Please write your question more clearly. I am having a hard time understanding what it >is that you are asking. > >_______________________________________________ >Xen-devel mailing list >Xen-devel@lists.xensource.com >http://lists.xensource.com/xen-devel_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Konrad Rzeszutek Wilk
2011-Apr-26 13:54 UTC
Re: Re: [Xen-devel] if hvm better than pv?why dd in hvm vm no larger than 50M, while pv greater than 50
On Tue, Apr 26, 2011 at 04:48:31PM +0800, yue wrote:> 1.i find pv disk''throughput is better than hvm. i want to know why?Pick up "The definitive guide to the Xen hypervisor" book.> 2.pci passthrough is said to improve performance, if it is default method of IO-process.Sure. But it also makes migration difficult.> 3.do you familar with ocfs2''s source code?No. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel