In my last two blogs we looked at technology selection recommendations for core configuration management database (CMDB) and application discovery and dependency mapping (ADDM) investments. In this blog I’ll present some of the criteria surrounding analytic and automation investments, with brief additional discussions of why visualization and reporting, social IT, project management, and service catalog offerings may also become important to your overall CMDB System success. As with the prior blogs, all the ideas and some of the texts are derived from “CMDB Systems: Making Change Work in the Age of Cloud and Agile”—as just a taste of the depth, content and insights contained in that book.
Why Analytics are Key
Enterprise Management Associates has done a great deal of research looking at the complementary benefits of combining service modeling with advanced analytics. In many cases analytic investments can benefit strongly from linkages to ADDM and/or CMDB/CMS solutions. Conversely, the CMDB System thrives when analytics bring added value by streamlining currency and administration, as well as enriching insights into changes and CI-related interdependencies.
Figure 1 shows a list of analytic priorities as highlighted by eleven CMDB vendors[1] based on under-the-cover or portfolio-driven analytics integrations (as opposed to analytics integrations with third-party sources). A brief discussion of these options should help to clarify why analytics are important.
If/then change impact analysis. Critical not only for managing change but also for minimizing service disruptions, this analysis allows you to view where and how a service might be disrupted, while adding insight into the impacts of changes already made on critical IT-delivered business services.
Most often associated with event management, correlation is important as a growing number of CMDB providers are enabling events to map to their service models to trigger actions—from diagnostics to change automation. Correlation may also be applied to analyzing CI-related data for more effective CMDB maintenance and accuracy, or for more quickly assessing the potential impacts of unapproved changes to the IT environment.
Anomaly/anomalous behaviors. These analytic capabilities are most often associated with performance and security-related management issues that may map directly to CMDB service modeling—triggering faster time to resolve issues, and generating significant OpEx efficiencies.
Predictive trending. Predictive trending is especially of value for performance capacity planning and optimization, IT governance, and optimizing the business of IT.
Data mining or On-Line Analytical Processing. This is often used as shorthand for flexible and powerful query-based reporting and trending, often with a Cognos backend. However, some CMDB vendors also integrate with true data mining tools at the back end.
Self-learning heuristics. Some CMDBs and ADDM investments integrate with analytics that can dynamically assess massive amounts of data, ‘learn’ normal behaviors based on time, and can trigger alerts or actions based on significant deviations. When mapped to service modeling in the CMDB System, these values can become especially powerful.
Not mentioned in the above list, but also relevant are Data Optimization Analytics. These analytics represent a new and growing market and provide an analytic framework for assimilating and reconciling ‘trusted sources’ including gaps in coverage. In some cases these solutions can even rank toolset efficiencies—e.g. Which inventory tool brings in the best coverage for Windows desktops? These analytics can be especially valuable when trying to assimilate what we’ve seen as more than 50 different sources for a CMDB System.
A Few Words on Automation
Automation is also key. One way to look at automation is to consider the core CMDB System as a primary investment in visibility—analytics included. Suddenly, you might say, the world is at your doorstep because you can see it all!
But then the question becomes, What do you want to do about it? The true value comes from unifying ‘visibility’ and ‘context’ with ‘action.’ Right now the industry is buzzing with vendors boasting about automation—but without visibility and context, as per an analytically enriched CMDB System, the result too often is automating train wrecks.
The top five most prevalent types of automation supported associated with CMDB Systems are—in ranked order:
Change and configuration management automation
ITIL process automation
Service desk trouble ticketing
Service catalog integrations (for service provisioning)
Service desk workflows
However, more integrated IT process automation, or runbook, is a growing area of investment as well, unifying these targeted analytic investments.
A look at some other investments
Here are a few other areas to consider in making an your CMDB System come alive with value:
Visualization and reporting: Visualization is the most demonstrable value as you can actually ‘see’ a vendor’s capabilities during proof of concept. One way to address visualization is to ask what roles and use cases are supported. For more general usage, some of the roles you should consider include ones that are domain specific: e.g., network, systems, desktop, mobile, security, service desk, applications, development. You need to also include roles that are cross-domain e.g., engineering, architecture, service delivery, change management, asset management, financial planning, user experience management. A third set of roles that you need to incorporate are ones that are enterprise/non-IT related: e.g., financial planning, non-IT executives, on-line operations, partner management, etc.
Social IT, Gamification, etc: This is another emerging technology area that can deliver benefits in terms of promoting better processes, commitment, and dialog among key stakeholders. Even as solutions in this area evolve, many are already providing strong values in a CMDB System context.
Project Management: While hardly limited to supporting a CMDB System, project management can be a strong plus, especially for managing and planning initial phase deployments. For CMDB deployments, you should seek out project management capabilities that are nimbler and less hierarchical than some of the more traditional brands.
Service Catalog (including for cloud): Ideally, a service catalog, or even a service portal, should be a natural extension of the CMDB System, with consistent modeling and insight into relevant interdependencies. This is just as true for those catalogs that include cloud-based services as it is for in-house delivered services, as it’s becoming increasingly important to understand dependencies outside IT’s physical walls.
Read more from Dennis on SciTech Connect:
Making the Right Application Discovery and Dependency Mapping (ADDM) Investment
CMDB Systems: Some Key (and Surprising) Findings from Deployments
Selecting the Right Technology for Your CMDB System, Pt. 1: The Core CMDB/CMS
[1] “CMDB/CMS Radar: the Move to Federation,” EMA, 2012.
CMDB Systems: Making Change Work in the Age of Cloud and Agile is available for purchase on the Elsevier Store.
Use discount code “STC215″ at checkout and save up to 30% on your very own copy!
About the Author
Dennis Nils Drogseth worked in technology for more than thirty years. He has been an analyst with Enterprise Management Associates for sixteen years, where he currently supports EMA through leadership in Business Service Management (BSM), CMDB Systems, automation systems and service-centric financial optimization. His extensive involvement in CMDB Systems dates back to 2004 where he pioneered unique research with distinctive industry insights during the formative years of the initial CMDB upsurge in the United States He has been widely published in columns and features in trade publications such as APM Digest, Network World and the CSC Executive Forum.
The post CMDB System Technology Selection Part Three: Why Analytics and Automation Need to Be a Part of the Game appeared first on SciTech Connect.