Jonathan LaCour, VP, Cloud and Development at DreamHost, a global web hosting, domain registrar and cloud services provider, discusses OpenStack from a user perspective…
Being a company with a core value to “Embrace Open Source,” when it came to launching a public cloud service it made sense to want that offering grounded upon powerful, versatile open source software that could deliver scalability, resiliency, and security. Because of that, it was an easy choice to base it upon OpenStack’s open source platform – and here’s why.
While the pre-cloud era enjoyed sunny weather for the open Internet – with wide use of free and transparent software such as Linux, Apache, MySQL and openly accessible programming languages such as PHP, Perl, Python, and Ruby – the arrival of closed source black box clouds (like Amazon Web Services and Microsoft Azure) have rained on open source’s parade and precipitated a push back against the transparency which developers, engineers, and IT organizations had enjoyed. With these cloud infrastructure services based on proprietary software, users have no knowledge on exactly how the most important component of their application stacks is being implemented.
Founded in 2010 through a somewhat surprising partnership between NASA and Rackspace Hosting, OpenStack has really enabled developers to create private and public clouds using open source cloud software, helping to shift the needle back towards open Internet transparency in the current cloud era.
Though not without its hiccups, OpenStack has since caught on like wildfire, with more than 500 technology companies joining the project and semi-annual OpenStack Summits continuing to experience exponential growth in attendance. Together, these companies and legions of individual developers are building open source software and APIs, used all over the world to enable every type of cloud.
The commitment to open source and OpenStack has paid dividends in the enthusiasm and appetite users have shown for highly capable cloud computing solutions based on open software. With our OpenStack-powered cloud, users create virtual machines, block devices, and networks on-demand, using standard OpenStack APIs. The architecture of OpenStack is modular, with components for separate functions. The networking services leverage OpenStack’s Neutron service, along with VXLAN (an open network virtualization standard), and the Akanda project. The dashboard is built with OpenStack’s Horizon component, which provides a graphical interface for accessing and provisioning cloud resources (command-line tools are available as well). Using OpenStack APIs, the service gives users the ability to establish system automation.
Fully-operational virtual machines in this particular example of an OpenStack-powered cloud platform can be created in as little as 40 seconds, and full network virtualization is provided as well. With the physical network representing an “underlay” invisible to the user, a virtual network is overlaid on top, equipping users with a virtual OSI Layer 2 switch, isolated for each individual user. Through OpenStack’s Neutron Networking APIs, this network virtualization technology gives users fully programmable control of their network from virtual OSI Layers 2-7, with total isolation. While this public cloud service does add some unique features (based, of course, upon additional open source software!), it maintains compatibility with the standard OpenStack Compute, Network, Image, and Storage APIs, ensuring that it is indeed a standard OpenStack deployment at its heart.
One of OpenStack’s promises is to really bring us back to a world with a more open Internet, where the inner workings of the primary cloud technology driving so many sites and services is transparent and can be understood and improved upon by all, often to the benefit of all. This is the spirit of the open source philosophy, and more than that, it’s also the backbone of the thriving OpenStack ecosystem, and a rich promise for the future of the open source cloud.