1 .. This work is licensed under a Creative Commons Attribution 4.0 International License.
2 .. http://creativecommons.org/licenses/by/4.0
5 ========================
6 Target software platform
7 ========================
12 This section will provide information which upstream projects, versions and components are
13 integrated in the Brahmaputra release
18 OPNFV uses OpenStack as cloud management system.
19 Brahmaputra is based on OpenStack Liberty Release. It comprises the following sub-projects
24 * Cinder (Block Storage)
25 * Swift (Object Storage)
26 * Ceilometer (Telemetry)
28 * Glance (Image Service)
29 * Heat (Orchestration)
32 Some of the sub-projects are not deployed in all scenarios.
33 Besides target software, also deployment and test framework use OpenStack components
34 (Fuel, Tempest, Rally)
40 OPNFV uses Linux on all target machines. Depending on the installers, different
41 distributions are supported.
43 Ubuntu 14 supported by Fuel, Compass and Joid installers
44 CentOS 7 supported by Apex and Compass
50 OPNFV Brahmaputra release supports three different SDN controllers:
52 * OpenDaylight (ODL, Beryllium release)
56 Depending on the SDN controller you are using, the featureset will vary.
57 Brahmaputra also provides scenarios without an SDN controller, just using OpenStack Neutron.
63 We need a high level paragraph here and a description of how we use ODL.
69 .. ONOS intro shortened from https://wiki.onosproject.org/pages/viewpage.action?pageId=2851517
71 ONOS stands for **O** pen **N** etwork **O** perating **S** ystem. ONOS provides the control plane
72 for a software-defined network (SDN), managing network components, such as switches and links,
73 and running software programs or modules to provide communication services to end hosts and
76 ONOS provides a platform for SDN applications and use cases for routing, management, or
77 monitoring services for software-defined networks.
79 ONOS can run as a distributed system across multiple servers, allowing it to use the CPU and
80 memory resources of multiple servers while providing fault tolerance in the face of server
81 failure and potentially supporting live/rolling upgrades of hardware and software without
82 interrupting network traffic.
84 The ONOS kernel and core services, as well as ONOS applications, are written in Java as bundles
85 that are loaded into the Karaf OSGi container. OSGi is a component system for Java that allows
86 modules to be installed and run dynamically in a single JVM.
88 More information on the internal design of ONOS may be found in
89 `User's Guide <https://wiki.onosproject.org/display/ONOS/User's+Guide>`_ and
90 `Architecture+Guide <https://wiki.onosproject.org/display/ONOS/Architecture+Guide>`_ on the
91 `wiki of the ONOS project <https://wiki.onosproject.org>`_.
93 ONOS is integrated to OPNFV using a framework ONOSFW and Neutron plugins. Details can be found in the
94 ONOS specific OPNFV documents:
103 We need a high level paragraph here and a description of how we use OpenContrail, including
104 its vRouter capabilities.
112 Deployment Architecture
113 =======================
115 OPNFV starts with a typical configuration with 3 controller nodes running
116 OpenStack, SDN, etc. and a minimum of 2 compute nodes for deployment of VNFs.
117 A detailed description of this 5 node configuration can be found in pharos documentation.
119 The 3 controller nodes allow to provide an HA configuration. The number of compute
120 nodes can be increased dynamically after the initial deployment.
122 OPNFV can be deployed on bare metal or in a virtual environment, where each of the hosts
123 is a virtual machine and provides the virtual resources using nested virtualization.
125 The initial deployment is done using a so-called "jumphost". This server (either bare metal
126 or virtual) is first installed with the installer program that then installs OpenStack
127 and other components on the controller nodes and compute nodes. See the installer
128 documentation for more details.
131 In a second level of detail, describe how software is distributed over the 3 controller
132 nodes, compute nodes and other hardware.
134 In Brahmaputra, the following scenarios are supported:
142 Editors note: we might skip this section completely for Brahmaputra.
144 Or we provide rather short statements. In later versions, we have to describe which
145 software is involved in which way during:
147 * VNF Life Cycle (onboarding, instantiate, scaling): we can reference to other documents
148 * Hardware Life Cycle (mainly how to add compute nodes, but also other cases)