[Link] in detail about OpenStack?
BTL4
(Definition:2 marks,Diagram:4 marks,Concept explanation: 9 marks)
The OpenStack project is an open source cloud computing platform for all types of
clouds, which aims to be simple to implement, massively scalable and feature rich.
Developers and cloud computing technologists from around the world create the
OpenStack project.
OpenStack provides an Infrastructure as a Service (IaaS) solution through a set of
interrelated services.
Each service offers an application programming interface (API) that facilitates this
[Link] on their needs, administrator can install some or all services.
OpenStack began in 2010 as joint project and 2012, it is managed by the OpenStack
Foundation, a non-profit corporate entity established in September 2013 to promote
OpenStack software and its community
Now, More than 500 companies have joined the project. The OpenStack system consists
of several key services that are separately installed.
These services work together depending on the user cloud needs and include the
Compute, Identity, Networking, Image, Block Storage, Object Storage, Telemetry,
Orchestration, and Database services.
The administrator can install any of these projects separately and configure them
standalone or as connected entities.
Figure 4.4 Relationship between OpenStack services, which illustrates how different
OpenStack components (modules) interact to provide cloud infrastructure as a service
(IaaS).
1. Keystone
Role: Provides authentication (auth) and authorization services for all OpenStack
components.
Function: Every service (Nova, Glance, Swift, etc.) uses Keystone to validate
credentials before granting access.
2. Ceilometer
Role: Monitors and measures usage.
Function: Collects telemetry data for billing, benchmarking, and scaling decisions.
3. Horizon
Role: Provides a User Interface (UI).
Function: A web-based dashboard (Graphical User interface- users can see a
webpage where they click buttons and fill forms) to manage OpenStack services
(launch instances, assign IPs, manage volumes, etc.).
4. Nova
Role: The compute controller.
Function: Manages the lifecycle of virtual machines (VMs).
o Boots data processing instances.
o Boots database instances.
o Provisions and manages VMs.
o Provides images to VMs (from Glance).
o Provides volumes to VMs (from Cinder).
o Connects to networks (via Neutron).
5. Glance
Role: Image service.
Function: Manages VM images.
o Registers guest images.
o Registers Hadoop images.(similar to VM image but mainly used for Bigdata
processing.)
o Provides/fetches images for Nova.
6. Neutron
Role: Networking service.
Function: Provides network connectivity for VMs.
o Allocates IP addresses.
o Manages networking topologies(structure for how different components are
connected together like routers, subnets, firewalls, VPN).
o Provides PXE boot networking (way to start a computer over network instead
of booting it from local hard disk, CD or USB)for bare-metal provisioning (via
Ironic – bare-metal provisioning service) for physical servers.
7. Swift
Role: Object storage service.
Function: Stores and retrieves unstructured data (objects, files).
o Stores VM images.
o Stores job binaries and data for Sahara.
o Backs up volumes from Cinder.
8. Cinder
Role: Block storage service.
Function: Provides persistent block-level storage (like hard drives).
o Provides volumes to VMs.
o Backs up volumes in Swift.
o Stores databases in Trove.
9. Ironic
Role: Bare-metal provisioning service.
Function: Provisions and manages physical (bare-metal) machines instead of VMs.
o Works with Neutron for PXE network boot.
10. Trove
Role: Database as a Service (DBaaS).
Function: Manages database instances (SQL, NoSQL).
o Boots database instances via Nova.
o Stores backups in Cinder.
11. Sahara
Role: Data processing service.
Function: Manages big data frameworks (pre-built structure)(like Hadoop, Spark).
o Orchestrates clusters for data analytics.
o Uses Swift to store job binaries/data.
o Uses Nova to boot data processing instances.
12. Heat
Role: Orchestration service.(automates the arrangement, coordination and
management of multiple resources so they work together as a single system)
Function: Automates cloud application deployment using templates.
o Provisions, operates, and manages resources (VMs, networks, volumes).
o Works with Sahara, Trove, and others for orchestration.
13. VMs (Virtual Machines)
Central to the diagram.
Provisioned and managed by Nova.
Receive:
o Images from Glance,
o Volumes from Cinder,
o Network from Neutron.
Act as the execution environment for user workloads.
How it all ties together:
1. User logs in via Horizon → authenticated by Keystone.
2. Nova provisions a VM using:
o Image from Glance,
o Volume from Cinder,
o Network from Neutron.
3. Data is stored in Swift (object) or Cinder (block).
4. Trove provisions databases, Sahara handles big data jobs, and Ironic provisions bare
metal servers.
5. Heat orchestrates everything as a single service deployment.
6. Ceilometer monitors for billing and scaling.
This diagram represents the ecosystem of OpenStack services and how they interconnect to
deliver IaaS cloud computing (compute, storage, networking, orchestration, and
management).
Step-by-Step VM Launch Workflow in OpenStack
Step 1: Authentication
User logs into Horizon (UI)
Horizon forwards credentials to Keystone.
Keystone authenticates and issues a token → used for all further requests.
Step 2: Request to Launch VM
User requests to launch a VM through Horizon (or CLI/API).
Horizon passes the request to Nova (compute service).
Step 3: Image Selection
Nova asks Glance for the requested image.
Glance fetches the image from its repository.
Image is sent back to Nova.
Step 4: Networking
Nova asks Neutron to assign:
o IP address,
o Networking (subnet, router, firewall rules).
Neutron provides the required network connection.
Step 5: Block Storage (Optional)
If persistent storage is needed:
o Nova requests a volume from Cinder.
o Cinder provisions the volume and attaches it to the VM.
Step 6: VM Provisioning
Nova now has:
o Image from Glance,
o Network from Neutron,
o Volumes from Cinder.
Nova provisions and boots up the VM.
Step 7: Data Storage (Optional)
If the user wants object/file storage:
o Data is saved in Swift (object storage).
o Volumes/DB backups can also be pushed into Swift.
Step 8: Advanced Services
Trove: If database instances are requested, Nova boots them using Cinder and Swift
for storage/backup.
Sahara: If big data processing is required, Sahara orchestrates clusters (Hadoop,
Spark).
Ironic: If user needs bare-metal servers instead of VMs, Ironic provisions them.
Step 9: Orchestration
Heat manages orchestration if the user deploys a multi-component application (e.g.,
web server + database + load balancer).
Heat automates provisioning, operation, and scaling.
Step 10: Monitoring
Ceilometer collects usage and performance data for billing, scaling, and monitoring.
Simplified Flow Diagram (textual)
User → Horizon (UI) → Keystone (Auth)
→ Nova (Compute)
→ Glance (Image)
→ Neutron (Network)
→ Cinder (Block Storage)
→ Swift (Object Storage)
→ Trove (Databases)
→ Sahara (Big Data)
→ Ironic (Bare Metal)
→ Heat (Orchestration)
→ Ceilometer (Monitoring)
To design, deploy, and configure OpenStack, administrators must understand the
logical [Link] consists of several independent parts, named the
OpenStack services. All services authenticate through a common Identity
[Link] services interact with each other through public APIs, except where
privileged administrator commands are necessaryInternally, OpenStack services are
composed of several processes.
All services have at least one API process, which listens for API requests,
preprocesses them and passes them on to other parts of the [Link] the exception
of the Identity service, the actual work is done by distinct [Link]
communication between the processes of one service, an AMQP message broker is
[Link] service's state is stored in a [Link] deploying and configuring the
OpenStack cloud, administrator can choose among several message broker and
database solutions, such as RabbitMQ, MySQL, MariaDB, and [Link] can
access OpenStack via the web-based user interface implemented by the Horizon
Dashboard, via command-line clients and by issuing API requests through tools like
browser plug-ins or [Link] applications, several SDKs are available. Ultimately, all
these access methods issue REST API calls to the various OpenStack services.
The controller node runs the Identity service, Image service, Placement service,
management portions of Compute, management portion of Networking, various
Networking agents, and the [Link] also includes supporting services such as an
SQL database, message queue, and NTP.
Optionally, the controller node runs portions of the Block Storage, Object Storage,
Orchestration, and Telemetry [Link] controller node requires a minimum of two
network [Link] compute node runs the hypervisor portion of Compute that
operates instances. By default,[Link]
[Link]. The compute node
also runs a Networking service agent that connects instances to virtual networks and
provides firewalling services to instances via security groups.
Administrator can deploy more than one compute node. Each node requires a
minimum of two network interfaces. The optional Block Storage node contains the
disks that the BlockStorage and Shared File System services provision for instances.
For simplicity, service traffic between compute nodes and this node uses the
management network.
Production environments should implement a separate storage network to increase
performance and security. Administrator can deploy more than one block storage
node. Each node requires a minimum of one network [Link] optional Object
Storage node contains the disks that the Object Storage service uses for storing
accounts, containers, and [Link] simplicity, service traffic between compute
nodes and this node uses the management [Link] environments should
implement a separate storage network to increase performance and [Link]
service requires two nodes. Each node requires a minimum of one network interface.
Administrator can deploy more than two object storage [Link] provider networks
option deploys the OpenStack Networking service in the simplest way possible with
primarily layer 2 (bridging/switching) services and VLAN segmentation of networks.
Essentially, it bridges virtual networks to physical networks and relies on physical
network infrastructure for layer-3 (routing) [Link], a DHCP service
provides IP address information to instances.