openstack/operations-guide
Code Issues Proposed changes

Addresses O'Reilly editor comments for Ch 4 Compute Nodes

Browse Source 
Change-Id: I8d1bca729eaa3d7ff0344d8af107bf50b8071d86
This commit is contained in:
Anne Gentle 2014-02-18 15:24:33 -06:00
parent dd116a8bc6
commit d138f75907
1 changed files with 14 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
doc/openstack-ops/ch_arch_compute_nodes.xml
View File

@ -40,6 +40,7 @@
those 12 cores are doubled to 24 cores. If you purchase a those 12 cores are doubled to 24 cores. If you purchase a
server that supports multiple CPUs, the number of cores is server that supports multiple CPUs, the number of cores is
further multiplied.</para> further multiplied.</para>
<sidebar><title>Multithread Considerations</title>
<para>Hyper-threading is Intel's proprietary simultaneous <para>Hyper-threading is Intel's proprietary simultaneous
multithreading implementation used to improve parallelization multithreading implementation used to improve parallelization
on their CPUs. You might consider enabling hyper-threading to on their CPUs. You might consider enabling hyper-threading to
@ -49,13 +50,15 @@
hyper-threading can be beneficial in intense computing hyper-threading can be beneficial in intense computing
environments. We recommend you do performance testing with environments. We recommend you do performance testing with
your local workload with both hyper-threading on and off your local workload with both hyper-threading on and off
to determine what is more appropriate in your case.</para> to determine what is more appropriate in your case.</para></sidebar>
</section> </section>
<?hard-pagebreak?> <?hard-pagebreak?>
<section xml:id="hypervisor_choice"> <section xml:id="hypervisor_choice">
<title>Hypervisor Choice</title> <title>Hypervisor Choice</title>
<para>OpenStack Compute supports many hypervisors to various <para>A hypervisor provides software to manage virtual machine access to
degrees, including: the underlying hardware. The hypervisor creates, manages, and
monitors virtual machines. OpenStack Compute supports many
hypervisors to various degrees, including:
<itemizedlist role="compact"> <itemizedlist role="compact">
<listitem><para><link xlink:title="reference manual" xlink:href="http://www.linux-kvm.org/">KVM</link> (http://www.linux-kvm.org/)</para></listitem> <listitem><para><link xlink:title="reference manual" xlink:href="http://www.linux-kvm.org/">KVM</link> (http://www.linux-kvm.org/)</para></listitem>
<listitem><para><link xlink:title="reference manual" xlink:href="http://lxc.sourceforge.net/">LXC</link> (http://lxc.sourceforge.net/)</para></listitem> <listitem><para><link xlink:title="reference manual" xlink:href="http://lxc.sourceforge.net/">LXC</link> (http://lxc.sourceforge.net/)</para></listitem>
@ -313,7 +316,6 @@
knowledge about it.</para> knowledge about it.</para>
</section> </section>
</section> </section>
<section xml:id="overcommit"> <section xml:id="overcommit">
<title>Overcommitting</title> <title>Overcommitting</title>
<para>OpenStack allows you to overcommit CPU and RAM on <para>OpenStack allows you to overcommit CPU and RAM on
@ -331,10 +333,10 @@
</itemizedlist> </itemizedlist>
<para>The default CPU allocation ratio of 16:1 means that the <para>The default CPU allocation ratio of 16:1 means that the
scheduler allocates up to 16 virtual cores per physical scheduler allocates up to 16 virtual cores per physical
core. For example, if a physical node has 12 cores, then core. For example, if a physical node has 12 cores, then the scheduler
192 virtual cores would be available and with typical sees 192 available virtual cores. With typical flavor definitions of
flavours, of 4 virtual cores per instance, this would 4 virtual cores per instance, this ratio would provide 48 instances
provide 48 instances on a physical node.</para> on a physical node.</para>
<para>The formula for the number of virtual instances on a <para>The formula for the number of virtual instances on a
compute node is <emphasis>(OR*PC)/VC</emphasis>, where: compute node is <emphasis>(OR*PC)/VC</emphasis>, where:
</para> </para>