Introduction & History
Mission Statement : To produce the ubiquitous Open Source Cloud Computing platform that will meet the needs of public and private clouds regardless of size, by being simple to implement and massively scalable.
· Founded in 2010
o Nebula platform (NASA)
o Cloud Files Platform (Rackspace)
· Commercial distributions since 2012
o Red Hat’s preview release on “Essex” 2012
o Red Hat’s commercially supported on “Grizzly” 2013
· NASA drops out in July 2013
o “Lack of technical progress”
Governance of OpenStack
· Non Profit foundation
o Board of 24 directors
8 platinum sponsors
8 elected out of the max. 24 gold sponsors
8 elected individual members
o Technical committee – The OpenStack Technical Committee provides technical leadership for OpenStack as a whole. – https://www.openstack.org/foundation/tech-committee/
o User committee – These are some of the guys using OpenStack within their environments – http://www.openstack.org/foundation/user-committee/
“The selection of OpenStack components has a significant direct impact on the overall design. While there are certain components that will always be present, (Compute and Image Service, for example) there are other services that may not need to be present. As an example, a certain design may not require the Orchestration module. Omitting Orchestration would not typically have a significant impact on the overall design however, if the architecture uses a replacement for OpenStack Object Storage for its storage component, this could potentially have significant impacts on the rest of the design.”
As stated above an OpenStack deployment does not need to contain all components but i guess these are the most commonly used components rather than core required. This is what i have at least discovered from my reading to create this post.
Block storage / Object storage
Central Directory of users, groups, roles
Mapped to services
Common authentication system
Can integrate with existing backend services
Supports multiple forms of authentication
Single registry catalogue
Core Component – #2 – Compute Service (nova)
Core fabric controller
Main part of the Iaas System
Designed to manage pools of resources
Works with widely available virtualisation technologies
Used eventlet, kombu (AMQP), SQLAlchemy…
Designed to scale horizontally on standard hardware
No proprietary hardware or software requirements
Ability to integrate with legacy systems and third party technologies
Core Component – #3 – Block Storage (cinder)
Persistent block level storage devices
Used with compute instances
Manages block devices for servers
Storage volumes fully integrated into nova
Self-service management using horizon
Supports a variety of storage platforms
Appropriate for performance sensitive scenarios
Databases. Expandable file systems, block storage
Core Component – #4 – Object Storage (swift)
Scalable redundant storage system
Objects and files are written to multiple disk drives
Spread throughout servers in the data centre
Data replication and integrity provided
Storage clusters scale horizontally
Simply add new servers
Uses software logic and inexpensive commodity hard drives and servers
Core Component – #5 – Networking (neutron)
Ensures the network is not a bottleneck
Gives users self-service ability
Provides different networking models
Manages ip addresses
Supports software-defined networking technologies (Cisco, VMware …. Lots more… )
High levels of multi-tenancy, massive scaling…
Provides an extension framework
For IDS, Load balancing, firewalls, VPNs….
Provides a graphical interface
Both for administrators and users
Access, provision and automate resources
Accommodates third party products and services
Billing, monitoring and additional management tools
Is brand able for service providers
Additional Services – #2 – Image Service (Glance)
Provides services for disk and server images
Does not store images, variations or instances
Catalogues them and holds their metadata
Backend used through glance
Including OpenStack object storage (swift)
API provides a standard REST interface
Provide live migration, stacked image templates etc….
Additional Services – #3 – Telemetry – (Ceilometer) (Metering)
Single point of contact for billing systems
Traceable, auditable, extensible
Additional Services – #4 – Orchestration – (Heat)
Orchestrates composite cloud applications
Using templates and REST APIs
There are also a number of projects i have not listed above because they are not ready yet…. these include
Trove – MySQL database as a service.
Sahara – Control of Hadoop clusters.
Ironic – Manage bare metal servers.
Macaroni – A messaging service
All of the above are still being developed by the OpenStack community and you may see this in upcoming releases.
There have been 10 releases of OpenStack since being founded in 2010, the codenames used for the OpenStack releases are voted on by the OpenStack Group. The Codenames are cities or counties near to where the corresponding OpenStack design summit takes/took place. However as you can see below the release of “Grizzly” which refers to an element of the state flag of California. The design summit took place in San Diego.
Juno is the most current version that was released on October 16th 2014, it contained almost 400 new features and lots of improvements.
One thing to also add is the frequency that OpenStack versions are being released, unlike many other distributions, vendors and other such technologies to release two a year is great going, i guess my question and i have not found the answer is will this continue to develop at such a frequent rate or will there be a decline on how many releases are made in 1 year.
The next release expected in April 2015 will most certainly be called “Kilo” following the alphabetical sequence we have seen so far in the release naming convention…..
Following April we should see the next release around November to co-inside with the 6 month release cycle. The OpenStack Summit this year will be held in Vancouver, Canada in May, the venue of the summit tends to have some input to the upcoming release. A few guesses could be
Liberty: Liberty is a village in the Canadian province of Saskatchewan
Lizard: The Lizard Range is a mountain range in British Columbia, Canada
London: London is a city in South-western Ontario, Canada
Love: Love is a small village in the Canadian province of Saskatchewan
My favourite would be Lizard and I think it will between Liberty and Lizard if these are on the short list, let’s wait and see.
The final section i wanted to include in this post is what distributions are available both free and commercially available with enterprise support.
Commercial OpenStack Distributions
Red Hat OpenStack Platform
Free OpenStack Distributions
Here are some of the resources i used when creating my post, i think the posts from Eric Wright especially the second take more of a deep dive into the OpenStack use cases and technologies and the final link is outlining “what could” come in the Kilo release in April 2015.