<div dir="ltr">In my case Centos 7 is using QEMU 1.5.3 ... which is <i>ancient</i>. This is on a node with a packstack install of OpenStack. If you have a different result, I would like to know why...<br><br>Got a bit further in my reading and testing. Also got my raw volume read performance in an instance from ~300MB/s (with some tweaking) up to the current ~800MB/s. Given the host raw volume read rate is ~1.2GB/s, and there are substantial improvements in the software stack (Linux, iSCSI, QEMU) in later versions ... this is a good result.<div><br></div><div>Found the main bottleneck was the iSCSI target in the physical Linux host. (Not in my current test case, in QEMU.) From online presentations on QEMU/iSCSI/Linux, there are known large improvements in more recent versions. Punt. Need to re-test on top of Ubuntu Trusty LTS (what most customers seem headed toward). Will re-base my testing, at some point.<br><br>My testing (for simplicity) is on an all-in-one node. <br><br>Curious what other folk are getting with very-fast iSCSI targets. What is the upper range?<br><br><br><br><br><br><br><br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Mar 7, 2016 at 7:59 AM, Chris Friesen <span dir="ltr"><<a href="mailto:chris.friesen@windriver.com" target="_blank">chris.friesen@windriver.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Just a heads-up that the 3.10 kernel in CentOS/RHEL is *not* a stock 3.10 kernel. It has had many things backported from later kernels, though they may not have backported the specific improvements you're looking for.<br>
<br>
I think CentOS is using qemu 2.3, which is pretty new. Not sure how new their libiscsi is though.<br>
<br>
Chris<span class=""><br>
<br>
On 03/07/2016 12:25 AM, Preston L. Bannister wrote:<br>
</span><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">
Should add that the physical host of the moment is Centos 7 with a packstack<br>
install of OpenStack. The instance is Ubuntu Trusty. Centos 7 has a relatively<br>
old 3.10 Linux kernel.<br>
<br>
From the last week (or so) of digging, I found there were substantial claimed<br></span>
improvements in /both/ flash support in Linux and the block I/O path in QEMU -<span class=""><br>
in more recent versions. How much that impacts the current measures, I do not<br>
(yet) know.<br>
<br>
Which suggests a bit of tension. Redhat folk are behind much of these<br>
improvements, but RHEL (and Centos) are rather far behind. Existing RHEL<br>
customers want and need careful, conservative changes. Folk deploying OpenStack<br>
need more aggressive release content, for which Ubuntu is currently the best base.<br>
<br>
Will we see a "Redhat Cloud Base" as an offering with RHEL support levels, and<br>
more aggressive QEMU and Linux kernel inclusion?<br>
<br>
At least for now, building OpenStack clouds on Ubuntu might be a much better bet.<br>
<br>
<br>
Are those claimed improvements in QEMU and the Linux kernel going to make a<br>
difference in my measured result? I do not know. Still reading, building tests,<br>
and collecting measures...<br>
<br>
<br>
<br>
<br>
On Thu, Mar 3, 2016 at 11:28 AM, Chris Friesen <<a href="mailto:chris.friesen@windriver.com" target="_blank">chris.friesen@windriver.com</a><br></span><span class="">
<mailto:<a href="mailto:chris.friesen@windriver.com" target="_blank">chris.friesen@windriver.com</a>>> wrote:<br>
<br>
On 03/03/2016 01:13 PM, Preston L. Bannister wrote:<br>
<br>
> Scanning the same volume from within the instance still gets the same<br>
> ~450MB/s that I saw before.<br>
<br>
Hmmm, with iSCSI inbetween that could be the TCP memcpy limitation.<br>
<br>
<br>
Measuring iSCSI in isolation is next on my list. Both on the physical<br>
host, and<br>
in the instance. (Now to find that link to the iSCSI test, again...)<br>
<br>
<br>
Based on earlier comments it appears that you're using the qemu built-in<br>
iSCSI initiator.<br>
<br>
Assuming that's the case, maybe it would make sense to do a test run with<br>
the in-kernel iSCSI code and take qemu out of the picture?<br>
<br>
Chris<br>
<br>
<br>
__________________________________________________________________________<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br></span>
<<a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a>><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><span class=""><br>
<br>
<br>
<br>
<br>
__________________________________________________________________________<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
<br>
</span></blockquote><div class="HOEnZb"><div class="h5">
<br>
<br>
__________________________________________________________________________<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
</div></div></blockquote></div><br></div>