[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Xen-devel] [PATCH RFC V4 0/5] kvm : Paravirt-spinlock support for KVM guests
- To: Avi Kivity <avi@xxxxxxxxxx>, Alexander Graf <agraf@xxxxxxx>
- From: Raghavendra K T <raghavendra.kt@xxxxxxxxxxxxxxxxxx>
- Date: Tue, 17 Jan 2012 00:18:38 +0530
- Cc: Jeremy Fitzhardinge <jeremy@xxxxxxxx>, Greg Kroah-Hartman <gregkh@xxxxxxx>, Gleb Natapov <gleb@xxxxxxxxxx>, linux-doc@xxxxxxxxxxxxxxx, Peter Zijlstra <peterz@xxxxxxxxxxxxx>, Jan Kiszka <jan.kiszka@xxxxxxxxxxx>, Virtualization <virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx>, Randy Dunlap <rdunlap@xxxxxxxxxxxx>, Paul Mackerras <paulus@xxxxxxxxx>, "H. Peter Anvin" <hpa@xxxxxxxxx>, Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx>, Xen <xen-devel@xxxxxxxxxxxxxxxxxxx>, Dave Jiang <dave.jiang@xxxxxxxxx>, KVM <kvm@xxxxxxxxxxxxxxx>, Rob Landley <rlandley@xxxxxxxxxxxxx>, X86 <x86@xxxxxxxxxx>, Ingo Molnar <mingo@xxxxxxxxxx>, Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>, Srivatsa Vaddagiri <vatsa@xxxxxxxxxxxxxxxxxx>, Sasha Levin <levinsasha928@xxxxxxxxx>, Sedat Dilek <sedat.dilek@xxxxxxxxx>, Thomas Gleixner <tglx@xxxxxxxxxxxxx>, Marcelo Tosatti <mtosatti@xxxxxxxxxx>, LKML <linux-kernel@xxxxxxxxxxxxxxx>, Dave Hansen <dave@xxxxxxxxxxxxxxxxxx>, Suzuki Poulose <suzuki@xxxxxxxxxxxxxxxxxx>, Glauber Costa <glommer@xxxxxxxxxx>
- Delivery-date: Wed, 18 Jan 2012 10:35:28 +0000
- List-id: Xen developer discussion <xen-devel.lists.xensource.com>
On 01/16/2012 07:19 PM, Avi Kivity wrote:
On 01/16/2012 03:43 PM, Raghavendra K T wrote:
Dbench:
Throughput is in MB/sec
NRCLIENTS BASE BASE+patch
%improvement
mean (sd) mean (sd)
8 1.101190 (0.875082) 1.700395 (0.846809) 54.4143
16 1.524312 (0.120354) 1.477553 (0.058166) -3.06755
32 2.143028 (0.157103) 2.090307 (0.136778)
-2.46012
So on a very contended system we're actually slower? Is this expected?
I think, the result is interesting because its PLE machine. I have to
experiment more with parameters, SPIN_THRESHOLD, and also may be
ple_gap and ple_window.
Perhaps the PLE stuff fights with the PV stuff?
I also think so. The slight advantage in PLE, with current patch would
be that, we are be able to say " This is the next guy who should
probably get his turn". But If total number of unnecessary "halt
exits" disadvantage dominates above advantage, then we see degradation.
One clarification in above benchmarking is, Dbench is run
simultaneously on all (8 vcpu) 3 guests. So we already have 1:3
overcommit when we run 8 clients of dbench. after that it was just
increasing number of clients.
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
