diff --git a/doc/common/ch_getstart.xml b/doc/common/ch_getstart.xml
index 5b675604b6..62a206e845 100644
--- a/doc/common/ch_getstart.xml
+++ b/doc/common/ch_getstart.xml
@@ -204,6 +204,7 @@
         <xi:include href="section_getstart_telemetry.xml"/>
         <xi:include href="section_getstart_orchestration.xml"/>
         <xi:include href="section_getstart_trove.xml"/>
+        <xi:include href="section_getstart_sahara.xml"/>
     </section>
     <section xml:id="feedback">
         <title>Feedback</title>
diff --git a/doc/common/section_getstart_sahara.xml b/doc/common/section_getstart_sahara.xml
new file mode 100644
index 0000000000..e66ff31b44
--- /dev/null
+++ b/doc/common/section_getstart_sahara.xml
@@ -0,0 +1,48 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<section 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="sahara-service">
+  <title>Data processing service</title>
+  <para>The Data processing service for OpenStack (sahara) aims to provide
+      users with simple means to provision data processing (Hadoop, Spark)
+      clusters by specifying several parameters like Hadoop version, cluster
+      topology, nodes hardware details and a few more. After user fills in
+      all the parameters, the Data processing service deploys the cluster in a
+      few minutes. Also sahara provides means to scale already provisioned
+      clusters by adding/removing worker nodes on demand.
+  </para>
+
+  <para>The solution addresses the following use cases:
+    <itemizedlist>
+      <listitem><para>Fast provisioning of Hadoop clusters on OpenStack for
+        development and QA.</para></listitem>
+      <listitem><para>Utilization of unused compute power from general
+        purpose OpenStack IaaS cloud.</para></listitem>
+      <listitem><para>Analytics-as-a-Service for ad-hoc or bursty analytic
+        workloads.</para></listitem>
+     </itemizedlist>
+  </para>
+
+
+  <para>Key features are:
+    <itemizedlist>
+      <listitem><para>Designed as an OpenStack component.</para></listitem>
+      <listitem><para>Managed through REST API with UI available as part
+        of OpenStack dashboard.</para></listitem>
+      <listitem><para>Support for different Hadoop distributions:
+        <itemizedlist>
+          <listitem><para>Pluggable system of Hadoop installation
+            engines.</para></listitem>
+          <listitem><para>Integration with vendor specific management tools,
+            such as Apache Ambari or Cloudera Management Console.</para></listitem>
+        </itemizedlist>
+      </para></listitem>
+      <listitem><para>Predefined templates of Hadoop configurations with
+        ability to modify parameters.</para></listitem>
+      <listitem><para>User-friendly UI for ad-hoc analytics queries based on
+        Hive or Pig.</para></listitem>
+    </itemizedlist>
+  </para>
+</section>
diff --git a/doc/install-guide/bk-openstack-install-guide.xml b/doc/install-guide/bk-openstack-install-guide.xml
index 6174fb694b..5224e914e1 100644
--- a/doc/install-guide/bk-openstack-install-guide.xml
+++ b/doc/install-guide/bk-openstack-install-guide.xml
@@ -219,6 +219,7 @@
     <xi:include href="ch_heat.xml"/>
     <xi:include href="ch_ceilometer.xml"/>
     <xi:include href="ch_trove.xml"/>
+    <xi:include href="ch_sahara.xml"/>
     <xi:include href="ch_launch-instance.xml"/>
     <xi:include href="app_reserved_uids.xml"/>
     <xi:include href="../common/app_support.xml"/>
diff --git a/doc/install-guide/ch_sahara.xml b/doc/install-guide/ch_sahara.xml
new file mode 100644
index 0000000000..dc45a13422
--- /dev/null
+++ b/doc/install-guide/ch_sahara.xml
@@ -0,0 +1,19 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<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_sahara">
+  <title>Add the Data processing service</title>
+      <para>The Data processing service (sahara) enables users to provide a
+          scalable data processing stack and associated management interfaces.
+          This includes provision and operation of data processing clusters as
+          well as scheduling and operation of data processing jobs.
+      </para>
+
+       <warning><para>This chapter is a work in progress. It may contain
+          incorrect information, and will be updated frequently.</para></warning>
+          <xi:include href="../common/section_getstart_sahara.xml"/>
+          <xi:include href="section_sahara-install.xml" />
+          <xi:include href="section_sahara-verify.xml" />
+</chapter>
diff --git a/doc/install-guide/section_sahara-install.xml b/doc/install-guide/section_sahara-install.xml
new file mode 100644
index 0000000000..707b014e86
--- /dev/null
+++ b/doc/install-guide/section_sahara-install.xml
@@ -0,0 +1,96 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<section 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="sahara-install">
+  <title>Install the Data processing service</title>
+  <para>This procedure installs the Data processing service (sahara) on the
+      controller node.</para>
+  <para>To install the Data processing service on the controller:</para>
+  <procedure>
+    <step os="rhel;centos;fedora;opensuse;sles">
+      <para>Install required packages:</para>
+      <screen os="rhel;centos;fedora"><prompt>#</prompt> <userinput>yum install openstack-sahara python-saharaclient</userinput></screen>
+      <screen os="opensuse;sles"><prompt>#</prompt> <userinput>zypper install openstack-sahara python-saharaclient</userinput></screen>
+    </step>
+    <step os="ubuntu;debian">
+      <warning><para>You need to install required packages. For now, sahara
+          doesn't have packages for Ubuntu and Debian.
+          Documentation will be updated once packages are available. The rest
+          of this document assumes that you have sahara service packages
+          installed on the system.</para></warning>
+    </step>
+    <step>
+      <para>Edit <filename>/etc/sahara/sahara.conf</filename> configuration file</para>
+      <substeps>
+        <step><para>First, edit <option>connection</option> parameter in
+          the <literal>[database]</literal> section. The URL provided here
+          should point to an empty database. For instance, connection
+          string for MySQL database will be:
+          <programlisting language="ini">connection = mysql://sahara:<replaceable>SAHARA_DBPASS</replaceable>@<replaceable>controller</replaceable>/sahara</programlisting>
+        </para></step>
+        <step><para>Switch to the <literal>[keystone_authtoken]</literal>
+         section. The <option>auth_uri</option> parameter should point to
+         the public Identity API endpoint. <option>identity_uri</option>
+         should point to the admin Identity API endpoint. For example:
+          <programlisting language="ini">auth_uri = http://<replaceable>controller</replaceable>:5000/v2.0
+identity_uri = http://<replaceable>controller</replaceable>:35357</programlisting>
+        </para></step>
+        <step><para>Next specify <literal>admin_user</literal>,
+          <literal>admin_password</literal> and
+          <literal>admin_tenant_name</literal>. These parameters must specify
+          a keystone user which has the <literal>admin</literal> role in the
+          given tenant. These credentials allow sahara to authenticate and
+          authorize its users.
+        </para></step>
+        <step><para>Switch to the <literal>[DEFAULT]</literal> section.
+          Proceed to the networking parameters. If you are using Neutron
+          for networking, then set <literal>use_neutron=true</literal>.
+          Otherwise if you are using <systemitem>nova-network</systemitem> set
+          the given parameter to <literal>false</literal>.
+        </para></step>
+        <step><para>That should be enough for the first run. If you want to
+          increase logging level for troubleshooting, there are two parameters
+          in the config: <literal>verbose</literal> and
+          <literal>debug</literal>. If the former is set to
+          <literal>true</literal>, sahara will
+          start to write logs of <literal>INFO</literal> level and above. If
+          <literal>debug</literal> is set to
+          <literal>true</literal>, sahara will write all the logs, including
+          the <literal>DEBUG</literal> ones.
+        </para></step>
+      </substeps>
+    </step>
+    <step><para>If you use the Data processing service with MySQL database,
+      then for storing big job binaries in sahara internal database you must
+      configure size of max allowed packet. Edit <filename>my.cnf</filename>
+      file and change parameter:
+      <programlisting language="ini">[mysqld]
+max_allowed_packet = 256M</programlisting>
+      and restart MySQL server.
+    </para></step>
+    <step><para>Create database schema:
+      <screen><prompt>#</prompt> <userinput>sahara-db-manage --config-file /etc/sahara/sahara.conf upgrade head</userinput></screen>
+    </para></step>
+    <step><para>You must register the Data processing service with the Identity
+      service so that other OpenStack services can locate it. Register the
+      service and specify the endpoint:
+      <screen><prompt>$</prompt> <userinput>keystone service-create --name sahara --type data_processing \
+  --description "Data processing service"</userinput>
+<prompt>$</prompt> <userinput>keystone endpoint-create \
+  --service-id $(keystone service-list | awk '/ sahara / {print $2}') \
+  --publicurl http://<replaceable>controller</replaceable>:8386/v1.1/%\(tenant_id\)s \
+  --internalurl http://<replaceable>controller</replaceable>:8386/v1.1/%\(tenant_id\)s \
+  --adminurl http://<replaceable>controller</replaceable>:8386/v1.1/%\(tenant_id\)s</userinput></screen>
+    </para></step>
+    <step><para>Start the sahara service:
+      <screen os="rhel;centos;fedora;opensuse;ubuntu;debian"><prompt>#</prompt> <userinput>systemctl start openstack-sahara-all</userinput></screen>
+      <screen os="sles"><prompt>#</prompt> <userinput>service openstack-sahara-all start</userinput></screen>
+    </para></step>
+    <step><para>(Optional) Enable the Data processing service to start on boot
+      <screen os="rhel;centos;fedora;opensuse;ubuntu;debian"><prompt>#</prompt> <userinput>systemctl enable openstack-sahara-all</userinput></screen>
+      <screen os="sles"><prompt>#</prompt> <userinput>chkconfig openstack-sahara-all on</userinput></screen>
+    </para></step>
+  </procedure>
+</section>
diff --git a/doc/install-guide/section_sahara-verify.xml b/doc/install-guide/section_sahara-verify.xml
new file mode 100644
index 0000000000..03440c4392
--- /dev/null
+++ b/doc/install-guide/section_sahara-verify.xml
@@ -0,0 +1,26 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<section 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="sahara-verify">
+  <title>Verify the Data processing service installation</title>
+  <para>To verify that the Data processing service (sahara) is installed and
+    configured correctly, try requesting clusters list using sahara
+    client.</para>
+  <procedure>
+    <step>
+      <para>Source the <literal>demo</literal> tenant credentials:</para>
+      <screen><prompt>$</prompt> <userinput>source demo-openrc.sh</userinput></screen>
+    </step>
+    <step>
+      <para>Retrieve sahara clusters list:</para>
+      <screen><prompt>$</prompt> <userinput>sahara cluster-list</userinput></screen>
+      <para>You should see output similar to this:</para>
+      <screen><computeroutput>+------+----+--------+------------+
+| name | id | status | node_count |
++------+----+--------+------------+
++------+----+--------+------------+</computeroutput></screen>
+    </step>
+  </procedure>
+</section>