[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] RE: [Xen-devel] Questions about xenoprof semantics
> -----Original Message----- > From: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx > [mailto:xen-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of > Jae-Wan Jang > Sent: Wednesday, September 06, 2006 10:37 PM > To: xen-devel@xxxxxxxxxxxxxxxxxxx > Subject: [Xen-devel] Questions about xenoprof semantics > > Hi, folks > > I am figuring out network bottleneck in domU using xenoprof. > > When I see the results, I can see the number of samples in > dom0 and domU differs. > > If the number of samples in domU is more than that in dom0, > can I safely assume that domU is running longer than dom0? > Yes, if you are using UNHALTED clock cycles as the hardware event, the total number of samples in each domain should be proportional to the number of busy clock cycles used in the domain context (including user level, kernel and xen code). If you see 2x the number of samples in one domain, it means its CPU utilization is also 2x. Renato Renato > Thanks in advance. > > > -- > > Jae-Wan Jang > jwjang@xxxxxxxxxxxxxxxxxx > +82-42-869-3559 > http://camars.kaist.ac.kr/~jwjang > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@xxxxxxxxxxxxxxxxxxx > http://lists.xensource.com/xen-devel > _______________________________________________ Xen-devel mailing list Xen-devel@xxxxxxxxxxxxxxxxxxx http://lists.xensource.com/xen-devel
|
![]() |
Lists.xenproject.org is hosted with RackSpace, monitoring our |