From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: Paravirtualization on VMware's Platform [VMI]. Date: Sat, 19 Sep 2009 10:43:40 +0300 Message-ID: <4AB48BAC.1000409@redhat.com> References: <1253233028.19731.63.camel@ank32.eng.vmware.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1253233028.19731.63.camel@ank32.eng.vmware.com> Sender: linux-kernel-owner@vger.kernel.org To: akataria@vmware.com Cc: Ingo Molnar , Thomas Gleixner , "H. Peter Anvin" , the arch/x86 maintainers , LKML , Chris Wright , virtualization@lists.osdl.org List-Id: virtualization@lists.linuxfoundation.org On 09/18/2009 03:17 AM, Alok Kataria wrote: > Hi, > > We ran a few experiments to compare performance of VMware's > paravirtualization technique (VMI) and hardware MMU technologies (HWMMU) > on VMware's hypervisor. > > To give some background, VMI is VMware's paravirtualization > specification which tries to optimize CPU and MMU operations of the > guest operating system. For more information take a look at this > http://www.vmware.com/interfaces/paravirtualization.html > > In most of the benchmarks, EPT/NPT (hwmmu) technologies are at par or > provide better performance compared to VMI. > The experiments included comparing performance across various micro and > real world like benchmarks. > We've reached a similar conclusion for kvm pvmmu vs ept/npt. -- Do not meddle in the internals of kernels, for they are subtle and quick to panic.