Systems Manageability – Part 2: Scope, Methodology and Ontology

  • Article

by kishi on April 06, 2007 04:14pm

In my last blog called “Why Manageability Matters” I talked about why we chose to work on “Systems Manageability” as a whole and get a grassroots understanding of it within the context of Linux and Open Source space. In this blog, I’m going to address the Methodology and Ontology of the Systems Manageability project. This will shed immediate light on how we approach, design and implement projects in the OSSL. Let’s start with the main goals and purpose behind the project.

I. Systems Manageability Project Goals:

  •  
    • Survey the scope, depth and breadth of Linux manageability solutions, primarily utilizing open source software
    • Use the knowledge from the research to develop an understanding of Manageability landscape in the Linux and Open Source
    • Build and simulate common Linux management solutions in the OSSL (SMORG scale – small and medium sized organizations )
    • Understand if a cohesive Linux manageability stack exists today--Understand manageability focus from a Linux Sysadmin’s perspective (SMORG scale)

Once we defined what we needed to get into, yet another realization dawned on us, which was the sheer size and volume of the data and information that was staring at us in the face. Let’s just say “overwhelmed” was a mild word compared to what we were looking at. My colleague, Steve Zarkos and I immediately realized that it was time to trim the scope of what we were doing and limit ourselves to what’s achievable in three months and with two people J. This called for drawing up what we considered to be “out-of-scope”, which was:

II. Out-of-scope:

  •  
    • Enterprise-level Management Stacks such as HP Openview, Tivoli etc.
    • Security and user management, which is a very large area in itself, and would include policies, permissions, authentication, identity management etc.
    • Asset management frameworks that allow for inventory management, asset tracking etc.
    • Disaster recovery and planning which would include backup technologies and data protection management

III. Systems Manageability Project Methodology:

The approach taken for the project was simple and scientific. The project was divided into three stages:

  •  
    • Stage 1 – Discovery and Definition: We would explore, uncover and scour the Linux Manageability landscape for toolsets, apps and frameworks prevalent in the SMORG scale workspace today
    • Stage 2 – Lab Implementation: We would implement the results of our research which would be actual Linux and Open Source products in the Lab
    • Stage 3 – Analysis and Conclusion: After discovering, defining and implementing Linux and Open Source based manageability toolsets, apps and frameworks in the lab, we would spend time understanding their form, function and experience to draw conclusions

IV. Systems Manageability Project Ontology (classification):

The hardest and most challenging aspect of the project was to develop some sort ontology, characterization or classification of the manageability technologies prevalent in the IT environments today.  The diagram below represents the overall "buckets" defined as part of this exercise.  Each section of the diagram is broken down to provide a detailed breakdown of each of these Systems Manageability classifications represented:

In the next blog to follow, I will break down the first segment of ontology i.e. “Provisioning and Deployment” and discuss our research with all of you. Meanwhile, we always look forward to hearing from you, our audience and urge you for any feedback you may have about the topic. Thank You for tuning into Port25.

Cheers!