IBM leverages OpenStack features;
SmartCloud Orchestrator benefits
by John Crawford
For the last few months, the recently renamed Cloud & Smarter Infrastructure Cloud Enablement team (formerly Tivoli Cloud Enablement) has been creating the training material for SmartCloud Orchestrator (SCO). As a part of the SCO team, I have attended meetings, created slides, written speaker notes, reviewed other people’s slides, commented on them and eventually as a team, conducted a T3 or “Train the Trainer” session for the entire department.
This is nothing new—it’s what our group does when there is a new cloud software offering or an update to IBM Smart Cloud Provisioning, Tivoli Service Automation Manager or IBM Service Delivery Manager. What was (or seemed) different with this offering is that it seemed to come together more quickly and more smoothly. I guess this could be because of the experience I have gained since I joined the department; it could be because this is a straightforward start to finish class, as opposed to our last venture in which each of the self-paced modules had to stand on its own; or it could be (and I choose to believe) that as a whole, the company’s architecture, development, testing and verification processes are maturing, and we’re getting better.
This is not to say (or admit) that we were doing things badly before, but SmartCloud Orchestrator is one of the first products I have been involved with from its inception that takes advantage of the OpenStack cloud environment and has been conceived, planned, and built with considerable emphasis on IBM’s Cloud Computing Reference Architecture (CCRA).
I am definitely not an expert on the CCRA, but in a nutshell, it is a framework for designing cloud computing offerings that has been developed by individuals representing divisions across all of IBM who have a depth of cloud technical, consulting and customer knowledge. CCRA looks at a cloud project from end to end, integrating business management functions with fully integrated management of cloud infrastructure and services. CCRA is used internally at IBM, as well as externally with customers. In the case of customers, it is used to design the customers cloud implementations.
Following the CCRA guidelines, developers and implementers can provide an open, secure solution that meets compliance for privacy and governance. SCO embodies these guidelines and provides Infrastructure as a Service (Iaas) and Platform as a Service (PaaS). When the guidelines are used to create a solution like SCO, it ends up being flexible, scalable and aims to reduce cost of ownership/implementation and increase return on investment (ROI).
A couple of paragraphs back, I mentioned that SCO was built to take advantage of OpenStack. For the uninitiated, OpenStack is a global collaboration of developers and cloud-computing technologists working to produce a ubiquitous IaaS open-source cloud computing platform for public and private clouds. In other words, a consortium of companies and individuals looked at the process of building a cloud infrastructure and figured out the things that every implementation had in common. When they had that list, they built modules that could be substituted for every implementation’s specific individual code. The modules are built to accept a fixed set of inputs and provide a fixed set of outputs through a communication layer that is easily adapted to specific needs.
The OpenStack group has developed “pluggable” modules for authentication , block storage, a user interface (dashboard), image management and virtual server deployment, to name a few.
Since every public or private cloud implementation requires the same basic functionality, having pre-packaged modules that can be plugged in reduces development time. Why should every cloud project re-invent the wheel? The modules are proven to work, so the developers can spend more time on the specific needs of a customer rather than coding, testing and tweaking an end to end solution. Compare it to making a sandwich. Rather than baking the loaf of bread, killing a turkey, cooking it, milking a cow and making cheese, you buy the bread, turkey and cheese and put it together with your favorite condiments.. You still end up with what you want, but you don’t have to do all the work yourself .
So, SCO is a unique IBM product for a couple of reasons. It uses OpenSource modules and it was designed from the beginning with the Cloud Computing Reference Architecture in mind. Those two things seem to have combined to create something entirely new and not to be dramatic, but revolutionary. I’m not (specifically) in sales, but, having seen and used SmartCloud Orchestrator, I’m proud to be associated with it. It has a learning curve, for sure. The Business Process Management capabilities are powerful and complex and will take some practice and brainpower, but nothing worth having and using comes easily. I think we’ve done something remarkable with SCO. Check it out. I think you’ll be impressed.
John Crawford is a Senior Enablement Specialist with IBM Cloud and Smarter Infrastructure Cloud Enablement. He has worked as a consultant, course developer and instructor for IBM since 1998. Reach him at firstname.lastname@example.org.