openstack-manuals/doc/admin-guide-cloud/ch_compute.xml
2015-01-24 10:04:32 +00:00

493 lines
28 KiB
XML

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE section [
<!ENTITY % openstack SYSTEM "../common/entities/openstack.ent">
%openstack;
]>
<chapter xmlns="http://docbook.org/ns/docbook"
xmlns:xi="http://www.w3.org/2001/XInclude"
xmlns:xlink="http://www.w3.org/1999/xlink"
version="5.0"
xml:id="ch_introduction-to-openstack-compute">
<title>Compute</title>
<para>The OpenStack Compute service allows you to control an
Infrastructure-as-a-Service (IaaS) cloud computing platform.
It gives you control over instances and networks, and allows
you to manage access to the cloud through users and
projects.</para>
<para>Compute does not include virtualization software.
Instead, it defines drivers that interact with underlying
virtualization mechanisms that run on your host operating
system, and exposes functionality over a web-based API.</para>
<section xml:id="section_system-architecture">
<title>System architecture</title>
<para>OpenStack Compute contains several main components.</para>
<para>
<itemizedlist>
<listitem>
<para>The <glossterm>cloud controller</glossterm> represents the global state
and interacts with the other components. The <literal>API server</literal>
acts as the web services front end for the cloud controller. The
<literal>compute controller</literal> provides compute server resources
and usually also contains the Compute service.</para>
</listitem>
<listitem>
<para>The <literal>object store</literal> is an optional component that provides
storage services; you can also instead use OpenStack Object Storage.</para>
</listitem>
<listitem>
<para>An <literal>auth manager</literal> provides authentication and
authorization services when used with the Compute system; you can also
instead use OpenStack Identity as a separate authentication service.</para>
</listitem>
<listitem>
<para>A
<literal>volume controller</literal> provides fast and
permanent block-level storage for the compute servers.</para>
</listitem>
<listitem>
<para>The <literal>network controller</literal> provides virtual networks to
enable compute servers to interact with each other and with the public
network. You can also instead use OpenStack Networking.</para>
</listitem>
<listitem>
<para>The
<literal>scheduler</literal> is used to select the
most suitable compute controller to host an
instance.</para>
</listitem>
</itemizedlist>
</para>
<para>Compute uses a messaging-based, <literal>shared nothing</literal> architecture. All
major components exist on multiple servers, including the compute, volume, and network
controllers, and the object store or image service. The state of the entire system is
stored in a database. The cloud controller communicates with the internal object store
using HTTP, but it communicates with the scheduler, network controller, and volume
controller using AMQP (advanced message queuing protocol). To avoid blocking a
component while waiting for a response, Compute uses asynchronous calls, with a callback
that is triggered when a response is received.</para>
<section xml:id="section_hypervisors">
<title>Hypervisors</title>
<para xlink:href="https://www.docker.io/">Compute controls hypervisors through an API
server. Selecting the best hypervisor to use can be difficult, and you must take budget,
resource constraints, supported features, and required technical specifications into
account. However, the majority of OpenStack development is done on systems using KVM and
Xen-based hypervisors. For a detailed list of features and support across different
hypervisors, see <link xlink:href="http://wiki.openstack.org/HypervisorSupportMatrix"
>http://wiki.openstack.org/HypervisorSupportMatrix</link>.</para>
<para>You can also orchestrate clouds using multiple
hypervisors in different availability zones. Compute
supports the following hypervisors:</para>
<itemizedlist>
<listitem>
<para><link xlink:href="https://wiki.openstack.org/wiki/Baremetal">Baremetal</link>
</para>
</listitem>
<listitem>
<para><link xlink:href="https://www.docker.io">Docker</link></para>
</listitem>
<listitem>
<para><link
xlink:href="http://www.microsoft.com/en-us/server-cloud/hyper-v-server/default.aspx"
>Hyper-V</link>
</para>
</listitem>
<listitem>
<para><link xlink:href="http://www.linux-kvm.org/page/Main_Page">Kernel-based
Virtual Machine (KVM)</link>
</para>
</listitem>
<listitem>
<para><link xlink:href="https://linuxcontainers.org/">Linux Containers (LXC)</link>
</para>
</listitem>
<listitem>
<para><link xlink:href="http://wiki.qemu.org/Manual">Quick Emulator (QEMU)</link>
</para>
</listitem>
<listitem>
<para><link xlink:href="http://user-mode-linux.sourceforge.net/">User Mode Linux
(UML)</link>
</para>
</listitem>
<listitem>
<para><link
xlink:href="http://www.vmware.com/products/vsphere-hypervisor/support.html"
>VMware vSphere</link>
</para>
</listitem>
<listitem>
<para><link xlink:href="http://www.xen.org/support/documentation.html">Xen</link>
</para>
</listitem>
</itemizedlist>
<para>For more information about hypervisors, see the <link
xlink:href="http://docs.openstack.org/juno/config-reference/content/section_compute-hypervisors.html"
>Hypervisors</link> section in the
<citetitle>OpenStack Configuration
Reference</citetitle>.</para>
</section>
<section xml:id="section_users-and-projects">
<title>Tenants, users, and roles</title>
<para>The Compute system is designed to be used by different
consumers in the form of tenants on a shared system, and
role-based access assignments. Roles control the actions
that a user is allowed to perform.</para>
<para>Tenants are isolated resource containers that form the
principal organizational structure within the Compute
service. They consist of an individual VLAN, and volumes,
instances, images, keys, and users. A user can specify the
tenant by appending <literal>:project_id</literal> to
their access key. If no tenant is specified in the API
request, Compute attempts to use a tenant with the same ID
as the user.</para>
<para>For tenants, you can use quota controls to limit the:</para>
<itemizedlist>
<listitem>
<para>Number of volumes that can be launched.</para>
</listitem>
<listitem>
<para>Number of processor cores and the amount of RAM that can be allocated.</para>
</listitem>
<listitem>
<para>Floating IP addresses assigned to any instance when it launches. This allows
instances to have the same publicly accessible IP addresses.</para>
</listitem>
<listitem>
<para>Fixed IP addresses assigned to the same instance when it launches. This allows
instances to have the same publicly or privately accessible IP addresses.</para>
</listitem>
</itemizedlist>
<para>Roles control the actions a user is allowed to perform.
By default, most actions do not require a particular role,
but you can configure them by editing the
<filename>policy.json</filename> file for user roles.
For example, a rule can be defined so that a user must
have the <parameter>admin</parameter> role in order to be
able to allocate a public IP address.</para>
<para>A tenant limits users' access to particular images. Each
user is assigned a user name and password. Keypairs
granting access to an instance are enabled for each user,
but quotas are set, so that each tenant can control
resource consumption across available hardware
resources.</para>
<note>
<para>Earlier versions of OpenStack used the term
<systemitem class="service">project</systemitem>
instead of <systemitem class="service"
>tenant</systemitem>. Because of this legacy
terminology, some command-line tools use
<parameter>--project_id</parameter> where you
would normally expect to enter a tenant ID.</para>
</note>
</section>
<section xml:id="section_storage-and-openstack-compute">
<title>Block storage</title>
<para>OpenStack provides two classes of block storage:
ephemeral storage and persistent volumes. Volumes are
persistent virtualized block devices independent of any
particular instance.</para>
<para>Ephemeral storage is associated with a single unique
instance, and it exists only for the life of that
instance. The amount of ephemeral storage is defined by
the flavor of the instance. Generally, the root file
system for an instance will be stored on ephemeral
storage. It persists across reboots of the guest operating
system, but when the instance is deleted, the ephemeral
storage is also removed.</para>
<para>In addition to the ephemeral root volume, all flavors
except the smallest, <filename>m1.tiny</filename>, also
provide an additional ephemeral block device of between 20
and 160&nbsp;GB. These sizes can be configured to suit your
environment. This is presented as a raw block device with
no partition table or file system. Cloud-aware operating
system images can discover, format, and mount these
storage devices. For example, the <systemitem
class="service">cloud-init</systemitem> package
included in Ubuntu's stock cloud images format this space
as an <filename>ext3</filename> file system and mount it
on <filename>/mnt</filename>. This is a feature of the
guest operating system you are using, and is not an
OpenStack mechanism. OpenStack only provisions the raw
storage.</para>
<para>Persistent volumes are created by users and their size
is limited only by the user's quota and availability
limits. Upon initial creation, volumes are raw block
devices without a partition table or a file system. To
partition or format volumes, you must attach them to an
instance. Once they are attached to an instance, you can
use persistent volumes in much the same way as you would
use external hard disk drive. You can attach volumes to
only one instance at a time, although you can detach and
reattach volumes to as many different instances as you
like.</para>
<para>You can configure persistent volumes as bootable and use
them to provide a persistent virtual instance similar to
traditional non-cloud-based virtualization systems.
It is still possible for the resulting instance to also have
ephemeral storage, depending on the flavor selected. In this
case, the root file system can be on the persistent volume
and its state maintained even if the instance is shut down.
For more information about this type of configuration, see
the <link
xlink:href="http://docs.openstack.org/juno/config-reference/content/">
<citetitle>OpenStack Configuration Reference</citetitle></link>.
</para>
<note>
<para>Persistent volumes do not provide concurrent access
from multiple instances. That type of configuration
requires a traditional network file system like NFS or
CIFS, or a cluster file system such as GlusterFS.
These systems can be built within an OpenStack cluster
or provisioned outside of it, but OpenStack software
does not provide these features.</para>
</note>
</section>
<section xml:id="instance-mgmt-ec2compat">
<title>EC2 compatibility API</title>
<para>In addition to the native compute API, OpenStack provides
an EC2-compatible API. This API allows EC2 legacy workflows
built for EC2 to work with OpenStack. For more information and
configuration options about this compatibility API, see the <link
xlink:href="http://docs.openstack.org/juno/config-reference/content/">
<citetitle>OpenStack Configuration Reference</citetitle></link>.
</para>
<para>Numerous third-party tools and language-specific SDKs
can be used to interact with OpenStack clouds, using both
native and compatibility APIs. Some of the more popular
third-party tools are:</para>
<variablelist>
<varlistentry>
<term>Euca2ools</term>
<listitem>
<para>A popular open source command-line tool for
interacting with the EC2 API. This is
convenient for multi-cloud environments where
EC2 is the common API, or for transitioning
from EC2-based clouds to OpenStack. For more
information, see the <link
xlink:href="http://open.eucalyptus.com/wiki/Euca2oolsGuide"
>euca2ools site</link>.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>Hybridfox</term>
<listitem>
<para>A Firefox browser add-on that provides a
graphical interface to many popular public and
private cloud technologies, including
OpenStack. For more information, see the <link
xlink:href="http://code.google.com/p/hybridfox/"
> hybridfox site</link>.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>boto</term>
<listitem>
<para>A Python library for interacting with Amazon
Web Services. It can be used to access
OpenStack through the EC2 compatibility API.
For more information, see the <link
xlink:href="https://github.com/boto/boto">
boto project page on GitHub</link>.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>fog</term>
<listitem>
<para>A Ruby cloud services library. It provides
methods for interacting with a large number of
cloud and virtualization platforms, including
OpenStack. For more information, see the <link
xlink:href="https://rubygems.org/gems/fog"
> fog site</link>.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>php-opencloud</term>
<listitem>
<para>A PHP SDK designed to work with most
OpenStack- based cloud deployments, as well as
Rackspace public cloud. For more information,
see the <link
xlink:href="http://www.php-opencloud.com">
php-opencloud site</link>.</para>
</listitem>
</varlistentry>
</variablelist>
</section>
<section xml:id="section_instance-building-blocks">
<title>Building blocks</title>
<para>In OpenStack the base operating system is usually copied
from an image stored in the OpenStack Image Service. This
is the most common case and results in an ephemeral
instance that starts from a known template state and loses
all accumulated states on virtual machine deletion. It is
also possible to put an operating system on a persistent
volume in the Cinder volume system. This gives a more
traditional persistent system that accumulates states,
which are preserved on the Cinder volume across the
deletion and re-creation of the virtual machine. To get a
list of available images on your system run:
<screen><prompt>$</prompt> <userinput>nova image-list</userinput>
<?db-font-size 50%?><computeroutput>+--------------------------------------+-------------------------------+--------+--------------------------------------+
| ID | Name | Status | Server |
+--------------------------------------+-------------------------------+--------+--------------------------------------+
| aee1d242-730f-431f-88c1-87630c0f07ba | Ubuntu 14.04 cloudimg amd64 | ACTIVE | |
| 0b27baa1-0ca6-49a7-b3f4-48388e440245 | Ubuntu 14.10 cloudimg amd64 | ACTIVE | |
| df8d56fc-9cea-4dfd-a8d3-28764de3cb08 | jenkins | ACTIVE | |
+--------------------------------------+-------------------------------+--------+--------------------------------------+</computeroutput></screen>
</para>
<para>The displayed image attributes are:</para>
<variablelist>
<varlistentry>
<term><literal>ID</literal></term>
<listitem>
<para>Automatically generated UUID of the
image</para>
</listitem>
</varlistentry>
<varlistentry>
<term><literal>Name</literal></term>
<listitem>
<para>Free form, human-readable name for
image</para>
</listitem>
</varlistentry>
<varlistentry>
<term><literal>Status</literal></term>
<listitem>
<para>The status of the image. Images marked
<literal>ACTIVE</literal> are available
for use.</para>
</listitem>
</varlistentry>
<varlistentry>
<term><literal>Server</literal></term>
<listitem>
<para>For images that are created as snapshots of
running instances, this is the UUID of the
instance the snapshot derives from. For
uploaded images, this field is blank.</para>
</listitem>
</varlistentry>
</variablelist>
<para>Virtual hardware templates are called
<literal>flavors</literal>. The default installation
provides five flavors. By default, these are configurable
by admin users, however that behavior can be changed by
redefining the access controls for
<parameter>compute_extension:flavormanage</parameter>
in <filename>/etc/nova/policy.json</filename> on the
<filename>compute-api</filename> server.</para>
<para>For a list of flavors that are available on your
system:</para>
<screen><prompt>$</prompt> <userinput>nova flavor-list</userinput>
<computeroutput>+-----+-----------+-----------+------+-----------+------+-------+-------------+-----------+
| ID | Name | Memory_MB | Disk | Ephemeral | Swap | VCPUs | RXTX_Factor | Is_Public |
+-----+-----------+-----------+------+-----------+------+-------+-------------+-----------+
| 1 | m1.tiny | 512 | 1 | 0 | | 1 | 1.0 | True |
| 2 | m1.small | 2048 | 20 | 0 | | 1 | 1.0 | True |
| 3 | m1.medium | 4096 | 40 | 0 | | 2 | 1.0 | True |
| 4 | m1.large | 8192 | 80 | 0 | | 4 | 1.0 | True |
| 5 | m1.xlarge | 16384 | 160 | 0 | | 8 | 1.0 | True |
+-----+-----------+-----------+------+-----------+------+-------+-------------+-----------+</computeroutput></screen>
</section>
<section xml:id="section_compute-service-arch">
<title>Compute service architecture</title>
<para>These basic categories describe the service
architecture and information about the cloud controller.</para>
<simplesect>
<title>API server</title>
<para>At the heart of the cloud framework is an API server,
which makes command and control of the hypervisor, storage,
and networking programmatically available to users.</para>
<para>The API endpoints are basic HTTP web services
which handle authentication, authorization, and
basic command and control functions using various
API interfaces under the Amazon, Rackspace, and
related models. This enables API compatibility
with multiple existing tool sets created for
interaction with offerings from other vendors.
This broad compatibility prevents vendor
lock-in.</para>
</simplesect>
<simplesect>
<title>Message queue</title>
<para>A messaging queue brokers the interaction
between compute nodes (processing), the networking
controllers (software which controls network
infrastructure), API endpoints, the scheduler
(determines which physical hardware to allocate to
a virtual resource), and similar components.
Communication to and from the cloud controller is handled
by HTTP requests through multiple API
endpoints.</para>
<para>A typical message passing event begins with the API
server receiving a request from a user. The API server
authenticates the user and ensures that they are permitted
to issue the subject command. The availability of objects
implicated in the request is evaluated and, if available,
the request is routed to the queuing engine for the
relevant workers. Workers continually listen to the queue
based on their role, and occasionally their type host name.
When an applicable work request arrives on the queue, the
worker takes assignment of the task and begins executing it.
Upon completion, a response is dispatched to the queue
which is received by the API server and relayed to the
originating user. Database entries are queried, added, or
removed as necessary during the process.</para>
</simplesect>
<simplesect>
<title>Compute worker</title>
<para>Compute workers manage computing instances on
host machines. The API dispatches commands to
compute workers to complete these tasks:</para>
<itemizedlist>
<listitem>
<para>Run instances</para>
</listitem>
<listitem>
<para>Terminate instances</para>
</listitem>
<listitem>
<para>Reboot instances</para>
</listitem>
<listitem>
<para>Attach volumes</para>
</listitem>
<listitem>
<para>Detach volumes</para>
</listitem>
<listitem>
<para>Get console output</para>
</listitem>
</itemizedlist>
</simplesect>
<simplesect>
<title>Network Controller</title>
<para>The Network Controller manages the networking
resources on host machines. The API server
dispatches commands through the message queue,
which are subsequently processed by Network
Controllers. Specific operations include:</para>
<itemizedlist>
<listitem>
<para>Allocate fixed IP addresses</para>
</listitem>
<listitem>
<para>Configuring VLANs for projects</para>
</listitem>
<listitem>
<para>Configuring networks for compute
nodes</para>
</listitem>
</itemizedlist>
</simplesect>
</section>
</section>
<xi:include href="compute/section_compute-images-instances.xml"/>
<xi:include href="compute/section_compute-networking-nova.xml"/>
<xi:include href="compute/section_compute-system-admin.xml"/>
<xi:include href="../common/section_support-compute.xml"/>
</chapter>