Microsoft To Update System Center Orchestrator 2012
Microsoft is planning some upcoming improvements to System Center Orchestrator 2012, as unveiled today at Microsoft's Worldwide Partner Conference in Los Angeles.
Orchestrator 2012 is Microsoft's workflow-process automation system for datacenters. Microsoft acquired the technology for Orchestrator when it bought Opalis Software in December of 2009. Since that time, the company has integrated the Opalis software to run on Windows Server 2008 and Windows Server 2008 R2 and has worked on rolling out integration packs for interoperability with various systems.
In June, Microsoft released a beta version of Orchestrator 2012. Now, the Orchestrator team has set its sights toward issuing a release candidate version, which is expected to arrive "in a few months' time," according to Adam Hall, a senior technical product manager at Microsoft, in a phone interview conducted last week.
"We are focusing on the major enterprise products out of the gate in terms of integration packs," Hall said. "We just released new versions of the VMware vSphere and IBM NetCool integration packs, which are two of the top used integration packs that we have. We are working on the HP integration packs as we speak as well."
Hall promised that Microsoft will have new integration packs (in beta form) for all of the System Center 2012 releases, including Operations Manager 2012, although he did not indicate when that would be. Microsoft announced its System Center 2012 products in March, but most of them were still at the beta testing phase at that time, with products expected to arrive sometime in the second half of this year.
"And we are also updating all of the existing System Center integration packs for the end-market products," Hall added, "so the 2007, 2008 and 2010 releases will get new integration packs with additional activities built in."
Hall said that Microsoft will have some updates to the Quick Integration Kit in Orchestrator 2012 "to make it even easier to create these integrations." In addition, the console for Orchestrator will be enhanced as it's still at the beta stage. Microsoft also plans to provide updated documentation to show how to do process automation (Opalis didn't offer much, Hall said). Orchestrator 2012 will be getting IPv6 support and lots of bug fixes, he added.
Microsoft is working to globalize Orchestrator 2012 for the release candidate version, which will mean it will be able to run on a non-English U.S. operating system, Hall said. He added that localization efforts will come later that will change the language used inside the application.
Orchestrator 2012 includes a designer that lets IT pros create runbooks through a graphical user interface modeler. A runbook is basically a list of common procedures for an IT environment. Orchestrator 2012 includes standard activities that are installed and available out of the box. Hall defined "activities" in Orchestrator 2012 as "individual tasks that perform a very defined action."
"The idea is that you link those together to create a runbook. As that runbook executes, it goes through activities, and you have looping and branching and all sorts of validations throughout the runbooks," he explained.
In contrast, integration packs are "build-on mechanisms," Hall said. Orchestrator 2012 includes 76 activities that can be used to connect to any runbook without using a single integration pack.
Users don't necessarily need an integration pack to talk to a system (such as Linux). That's a common misconception, Hall said.
"Certainly, out of the box, we can talk to just about anything and create a runbook of just about any variety," he explained.
The databus is Orchestrator 2012's publish and consume mechanism. Hall said that the databus "is really the heart and soul of what Opalis was built on and what we've leveraged for Orchestrator."
The databus can help create "a rich set of information that enables a very powerful runbook to be created without needing to know how to go get that piece of information," Hall explained. Outputs get put on the databus and are then passed on to subsequent activities, creating a cumulative effect.
"When the first activity executes in a runbook, it takes some inputs to get it going, and then it does something and creates an output," Hall explained. "That output is put on the databus and is passed on to all subsequent activities. And so, as a runbook executes and progresses, you get more and more information on this databus. And each subsequent activity is able to leverage that, manipulate and work with that information, add more onto it and then pass it on to the next activity in the runbook."
Kurt Mackie is online news editor for the 1105 Enterprise Computing Group.