James Houghton

Subscribe to James Houghton: eMailAlertsEmail Alerts
Get James Houghton via: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn

Related Topics: Cloud Computing, Virtualization Magazine, Blueprint4IT, Cloud Application Management


CMDBs in the Cloud

IT organizations will adopt comprehensive IT Blueprinting to complement CMDBs to give their Ops teams a fighting chance

As enterprises move inexorably towards the Cloud, a faint murmur in the distance is growing louder and louder. What is it - the security forces objecting...the CFO wondering where their money is going now...business teams clamoring for faster response time? Good guesses, but this group never makes quite that much noise - it's the operational teams scratching their heads wondering how, particularly given their existing set of challenges, they are going to keep their heads above water trying to manage something as amorphous as a Cloud.

They are right to be concerned. After all, when the glamour and excitement of deploying a new Cloud application/environment has passed and the industry fervor shifts to the next Big Thing, they will be the ones left with the responsibility to ensure that business expectations are met for years to come. But just how are they to do that? While impressive advancements have been made with respect to performance monitoring, capacity management, and provisioning, they have mostly been made in independent silos. Although these improvements have provided tangible benefits, it has required human intelligence to integrate and manage. Unfortunately we all know humans make mistakes, and are not very scalable.

In years past we talked of a magical solution to these woes in the form of a Configuration Management Database (CMDB)...a single source of the truth for all things IT, where desperate operations or engineering staff can go for the answer to all of their questions like the oracle of Delphi of ancient times (no relation to Mr. Ellison). As many found out - including this author - the promise of the CMDB greatly exceeded reality.

In recent months the topic has had a renaissance as many of the challenges of managing a virtualized environment have grown and been exacerbated by the adoption of Cloud. Unfortunately the crux of these discussions has not been about new solutions, but rather about new problems that confound a CMDB. These challenges include, but are not limited to:

  • An enormous amount of new data and relationships, as Cloud and virtual environments present new configuration items (CIs) that must be captured in the taxonomy of the CMDB. In our ‘legacy' days a server was a server; now a robust server can host several applications, complete with virtual routers/switches, web servers, firewalls, application servers, and database servers. The complexity of these new relationships exacerbates troubleshooting, as operations teams struggle to isolate and remediate performance issues of separate applications sharing common infrastructure.
  • A highly accelerated rate of change, as applications can be spun up and down in minutes, given more or less capacity, or migrated across the room or to an alternate data center with minimal, if any, administrator interaction.
  • Forced interaction of different teams and processes in the IT organization. Is the addition of capacity to a workload demanding new resources driven by traditional operations support doing change management, performance management, the capacity management team, or the team responsible for provisioning? Tough to tell when the whole process could be automated and take 30 seconds, but one thing is for sure: all of the teams are going to want to keep track of those changes.
  • Business Intelligence for IT - not to coin a new term, but simply to point out that the primitive search capabilities in current CMDBs are far from robust. It is very difficult to find the answers you may be looking for unless you know exactly what to query (in which case you probably don't need the CMDB in the first place).

Others before have made these points; the objective in raising them is not to add another voice to the din but instead to add a divergent perspective. The incumbent vendors (IBM, BMC, CA, et al) will undoubtedly define new data structures and embed new capabilities to address most of the challenges outlined above. But no matter what they do, with the current approach a CMDB can never be fast enough. The rate of change isn't the problem; it's a symptom - the speed of Cloud and virtualization just make it obvious. The true problem is that our entire approach to CMDBs is from an operational perspective, so it will always be reactive.

What's a frustrated Ops person to do? The answer lies not in a CMDB, but rather in augmenting a CMDB with additional perspective: demand characterization.

Demand characterization is not about simply documenting a set of service levels the business requests. Instead it is about understanding the nature of the business activities that an application fulfills, its importance relative to other systems, and understanding the conditions under which it is entitled to consume additional resources - and conversely have them taken away. This information should be codified into a comprehensive IT Blueprint for that system, one that understands the abstraction of workloads from physical resources, one that lives long past the design and build phase, and is consulted throughout the entire life cycle of that application. This information is also the single source for building the policy for the application that allows for dynamic change to occur. In this manner the CMDB with the last-known configuration and the Blueprint with the acceptable automated change parameters provides the information an operations team needs to maintain an application, whether hosted on an internal VM or an external Cloud provider.

In summary, CMDBs are hardly a prehistoric relic; the need is very real for understanding the last known configuration, for maintaining the ‘golden image,' for meeting audit requirements, and certainly to aid in financial management. But in our brave new world of Cloudy, virtualized environments it's not enough. Successful IT organizations will adopt comprehensive IT Blueprinting to complement CMDBs to give their Ops teams a fighting chance.

More Stories By James Houghton

James Houghton is Co-Founder & Chief Technology Officer of Adaptivity. In his CTO capacity Jim interacts with key technology providers to evolve capabilities and partnerships that enable Adaptivity to offer its complete SOIT, RTI, and Utility Computing solutions. In addition, he engages with key clients to ensure successful leverage of the ADIOS methodology.

Most recently, Houghton was the SVP Architecture & Strategy Executive for the infrastructure organization at Bank of America, where he drove legacy infrastructure transformation initiatives across 40+ data centers. Prior to that he was the Head of Wachovia’s Utility Product Management, where he drove the design, services, and offering for SOA and Utility Computing for the technology division of Wachovia’s Corporate & Investment Bank. He has also led leading-edge consulting practices at IBM Global Technology Services and Deloitte Consulting.

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.