Lessons Learned from OpenStack
OpenStack was going to create an open source alternative to AWS, as well as displace VMware in the Enterprise. But change is never easy, and some things didn't go as planned. Looking back, what can we learn from the lessons of OpenStack?
SHOW SPONSOR LINKS:
- Eucalyptus Cloud (2006)
- CloudStack (2008)
- History of OpenStack
- OpenStack Projects
- OpenStack accepting VMware was a Mistake (Mirantis, 2012)
- The trouble with OpenStack (Simon Wardley, 2013)
- OpenStack’s future depends on embracing AWS, Now (Cloudscaling, 2013)
- OpenStack adopts the “Big Tent” model (2015)
- OpenStack Summit 2016 (Austin) - The focus moves to Telco
HOW DID OPENSTACK EVOLVE?
An open source, programmable cloud infrastructure. Eucalyptus and CloudStack existed prior to OpenStack. Created as a partnership between NASA (Nova) and Rackspace (Swift).
API-driven, software-defined infrastructure
Every vendor (HW and SW) was on-board: the anti-VMware project
OpenStack Foundation was created - control was maintained by Rackspace (initially).
Focus was split on building an AWS-alternative &/or a VMware-alternative
Was it an integrated platform, or independent projects, or a combination?
What defined something as “OpenStack”? How do you upgrade OpenStack?
The scope eventually got too big - way behind an IaaS (PaaS, DBaaS, Hadoop, etc.)
- Cloud Foundry, OpenShift launched in 2011
- Threats from Hadoop companies
VMware admins didn’t know how to use python. AWS admins wanted AWS APIs
Where were the customers?
LESSONS LEARNED FOR THE FUTURE
- Independent governance body; leverage the Linux foundation (events, infrastructure, etc,)
- Focus on enabling something new (cloud-native apps, more flexible than PaaS), not replacing something existing
- Enable early customers, and remain customer focused.
- Email: show at thecloudcast dot net
- Twitter: @thecloudcastnet