A recent post from Doug Davis “Docker: Managing the excitement” has some nice insights on Docker as a “relatively new” technology. Among the other Docker niceties that Doug points out, one that resonated with me is, and I quote “it takes seconds (if not less than a second, after the first time since things are cached) for it to complete”. Reading between the lines in some of my other posts, I’ve often battled the dark forces of virtual machine world, usually trying to squeeze every bit of extra juice out of my poor laptop. Then I tried Docker and well, suffice to say I need fewer coffee breaks.
My long-time colleague and friend Takehiko Amano has a post on “Immutable Infrastructure” where he talks about Docker in the context of OpenStack and with a bunch of my earlier posts focused on OpenStack, I decided to figure out how the three – OpenStack, Docker, and UCD+P might work together.
One of the neat features of UrbanCode Deploy with Patterns (UCDP) is the ability to deploy blueprints to multiple clouds. Last time around I wrote about hooking up UCDP to an OpenStack cloud. In this post I’ll look at how to setup UCDP to connect to Amazon EC2.
UrbanCode Deploy with Patterns (UCDP): a mouthful right? A pretty neat and powerful mouthful though. Consider theses two posts: Cooking up deployments on stacked clouds and UrbanCode Deploy with vSphere and Chef. UCDP makes it simpler and easier to carry out a bunch of steps detailed in those posts. Before I get into the details, let me first quickly go through how to install UCDP and connect it to an OpenStack setup.
Remember that old,old post of mine: HTTP Filtered Value Set Providers and dependent enumerations ? Probably not. It goes into a little feature of Rational Team Concert that makes it easy to avoid having to duplicate data from external systems that RTC might need to integrate with. UrbanCode Deploy can do something similar for value sets for process properties.