Lately I have been hearing the statement, “I no longer need configuration management anymore – we’re moving to the Cloud” and I struggle to understand what this means. I wonder if this was a common statement back in the early 2000’s when virtualization was first introduced and becoming common-place in the enterprise.
For me, the platform used to provision servers and software is irrelevant to the question of “Do I need configuration management”. At the core of it all, configuration management really has nothing to do with the platform, or even the tools used to collect and manage the configuration data. It comes down to the processes that surround configuration management that are important for you to run your business.
As you consider the role of configuration management, think about these questions
- Do I have all the data I need, and from every platform, to support my configuration management business processes?
- Who can create an application, and what are all the components of that application?
- What will happen to my services if a software application is stopped?
- Who needs to be notified when an application has planned downtime?
- Is there sufficient information attached to my servers, software and applications that I can properly support them when something is broken?
- Who owns the job of ensuring the details of all my enterprise’s applications are current and up-to-date?
- If I am audited, do I have details of the software installed on all my servers and workstations?
For each of these questions, the statement of “we’re moving to the cloud” has no relevancy, yet these questions are still core to the overall process of Configuration Management.
What is important during this time is having a tool that can be deployed to support your specific configuration management business processes. It must have sufficient capability to meet any business need that arises, yet flexible enough that it can be configured to meet those needs. You need a tool that can collect and integrate data from any data source, under any security model, into a single view of that data. This is vital so you can be confident your decisions accurately reflect both the business need and the current state of your infrastructure.
The tools provided by the cloud vendors will provide you with the ability to provision and manage your infrastructure. They don’t provide the ability to apply business processes around that infrastructure. This is where Universal CMDB and Universal Discovery become relevant.
Universal Discovery (UD) provides the tools to discover and collect information from any data source. They can discover the existence and configuration of physical or virtual servers you have in your own datacenters, or all of your cloud-based compute resources. UD can discover using OS-based tools or by talking directly with cloud management tools exposed by AWS, Azure and Google. UD can discover your servers and software using agents—if your security teams won’t allow you to deploy a single account to all your computing systems. Or UD can discover via service accounts deployed to your systems without the need for any remote agent.
In the case of our latest release, CMS 11, Universal Discovery can now natively discover your full ecosystem on AWS with a single discovery job, including all of the software installed and in use on your EC2 compute instances. Universal Discovery can now also capture your cloud environments from the context of the VPNs you’ve deployed, and in the case of AWS, in the context of CloudFormation and Auto-Scaling groups you have set up.
All of this data is sent to a single repository we call Universal CMDB (UCMDB). We call it Universal because it captures information regardless of the source of the information. In addition to discovery, you may need to pull in data from other tools such as Micro Focus’s network management tool—Network Node Manager i—or an asset management tool such as Micro Focus’s Asset Manager. Or perhaps you’re still using spreadsheets in some cases – that data can be loaded into UCMDB as well.
This is important because to answer the seven questions above requires a single view of your data. Without this comprehensive, single view, you cannot be confident that any answers you provide will sufficiently cover your full infrastructure. And as a Configuration Management Process Owner, you need to know when someone brings down an application or database to perform maintenance that it won’t unexpectedly bring down your whole ecosystem. No one wants to receive that phone call!
Make sure you are prepared for whatever comes your way by truly understanding your data with Universal Discovery and Configuration Management.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.