The latest version of Puppet Labs’ popular IT automation tool features a rewritten orchestration engine that it says will ease the automated rollout of changes across multiple sites and clouds.
A new centralized database for all the data Puppet collects about a company’s deployments gives devops a full picture of what’s really going on in their world. PuppetDB was written in Clojure language and enables the new orchestration engine to provide much more granular control of rollouts, said Nigel Kersten, CTO of Puppet Labs (see disclosure).
“Puppet collects massive amounts of information to do its job — it knows about all the nodes that are running so we can be scalpel-like in our precision. We can pick a subset of machines to run, maybe just those in the Pacific time zone or those running AMD CPUs that match a certain profile,” he said. “Nodes are much more ephemeral these days [than in the past], I can’t remember the last time I saw a CMDB [configuration management database] deployed. We can ask the actual nodes what’s going on.”
That means that new Puppet can facilitate progressive deployments. A devops team can thus just roll out a new configuration on Amazon U.S. West to make sure everything goes well and then schedule rolling updates across the rest of Amazon Web Services, he said.
The goal of Puppet Enterprise 3.0 is to enable the configuration and roll out of (buzzword alert) software-defined infrastructure. “You can bundle your compute, network and storage configurations all in one place and then move all that stuff around as needed,” Kersten said.
Puppet Labs and rival Opscode are the leaders in configuration management tools for devops, but other players are emerging. For example, Saltstack, a Salt Lake City-based startup, for example, just won GigaOM’s Structure Launchpad competition.
Disclosure: Puppet Labs is backed by True Ventures, a venture capital firm that is an investor in the parent company of this blog, Giga Omni Media. Om Malik, founder of Giga Omni Media, is also a venture partner at True.