Efficiency Approach


Taking an efficiency approach into process improvement and moving up the OMCM stack is one way of analyzing your set of solutions. By knowing where you are in the people, process, and tools scale, you can understand what your organization is and is not capable of undertaking. Changes designed to improve lower-scoring areas should also free up sources of productivity and revenue because most analyses of data centers find that 80 percent of costs are non-hardware related. Therefore, do not just look to the technology portion of your solution for cost savings and efficiency.

By using standard industry-accepted frameworks for the efficiency methodology analysis, the stigma of judgment is often replaced with acceptance of the methodology. In addition, the categories, scoring, and tasks to improve become more easily understood and reportable, and people might already have knowledge of ITIL, which they can bring into the discussion. Because there are probably many people in your organization who already know where the low-hanging fruit exists, using a solid framework simply provides everyone a means to focus on organizing to extract those efficiencies.

Regardless of the method used, the previous chapters demonstrated how important it is to have a clear picture of what you currently do before deciding where to implement N1 Grid technology. Do not get too caught up in your actual score. The scoring is a tool to focus attention and to compare against after N1 Grid system action has been taken. In "Operational Management Capabilities Model" on page 74, some of the aspects of the categories used to score and the levels of maturity to use in the model are discussed. The maturity level scoring ranges from 1 to 5, with 1 indicating a level characterized by crisis control and 5 being an organization focused on business value management.

Chaos Resolution With Efficiency

OMCM level 1 to 2 activities should focus on resolving chaos with efficiency and reaching a state of efficient component management. This section contains best practices on how to analyze your N1 Grid solutions to resolve chaos with efficiency.

One definition of insanity is to keep doing the same thing over and over and expect a different result. Chaotic environments would benefit the most from N1 Grid solutions, but they are too often environments in which the IT staff does not have the time to organize and write down what they do because they spend all of their time working hard to keep the environment running. This lack of organization and process is exactly why the environment is in this state in the first place. Installations are often a one-off with very little knowledge capture to leverage, and few processes are defined, much less automated.

Small steps such as the following can help:

  • Have standard forms for service requests that are filled out online.

  • Begin to leverage technologies like Solaris JumpStart or Java QuickStart to automate the operating system or application provisioning.

  • Begin to reduce the number of disparate hardware, operating system, and application versions in the organization.

Chaotic environments often have few tools to provide visibility into the health and trends of the data center they manage. Problems found by end users take time to resolve, and little correlation or tracking of problems is performed. Data is crucial to break the cycle of reactivity. For instance, you should:

  • Identify the complete list of components in the IT infrastructure.

  • Begin a simple log that tracks service events and their resolution.

  • Begin to install and use some type of enterprise observability tool for compute, network, and storage resources that enables threshold and service event alarms to be centrally reported and logged.

  • Use a checklist of server installation steps as a baseline for how power, rack space, IP addresses, compute, network, and storage resources are requested and how operating systems and applications are installed.

  • Review architecture and production life cycle methodologies, and design a method for use in your organization.

For level 1 organizations, solutions that do not support these goals or require support levels greater than these goals should be re-architected against the gaps they create.

Proactive Organization

OMCM level 2 to 3 activities should focus beyond efficient component management towards proactive organization. This section contains best practices for analyzing your N1 Grid solutions to get more proactive.

It takes work to move beyond basic visibility and control for individual components to a level where it is possible to anticipate problems or reduce their impact through faster resolution. To move towards a proactive state, you should:

  • Focus on the management of the operational aspects of IT.

  • Categorize and prioritize end-user applications for allocation of IT resources in a more efficient manner.

  • Make service delivery more formal and repeatable through the use of the methodologies that are chosen for your organization.

Proactive test strategies for application sizing and runtime environment assumptions are the best way to reduce the impact of problems by avoiding them the first place. The ability to rapidly recover must be coupled with an understanding of what is running in the environment. Thorough and repeatable testing enables you to:

  • Size and track growth curves.

  • Set accurate thresholds to enable proactive reporting of out-of-band behavior.

  • Revisit assumptions or test bug fixes when problems are encountered.

For level 2 organizations, solutions that do not support these goals or require support levels greater than these goals should be re-architected against the gaps they create.

Service Management Efficiency

OMCM level 3 to 4 activities should focus on linking IT closer to the business and providing and using information for service management. This section contains best practices for analyzing your N1 Grid solutions to provide and use information.

Level 3 organizations can leverage their test harnesses and service delivery technologies into an organization with predictable and repeatable service delivery capability. With a solid architecture, you can work with mobile decomposed entities in organizations with consistent and measurable IT processes. To achieve this, you should:

  • Ensure that what is running in the production environment is exactly what was tested in the test environment.

  • Have the capabilities to move entities around, as needed, in any of their development, testing, or production environments.

  • Ensure that the participants in your architecture and production life cycle methodologies continue to expand beyond your IT organization into the business units and to plan instrumentation for business-level optimization of resources.

  • Ensure that your IT organization manages itself by SLAs, both internally and externally.

  • Ensure that business units and the IT organization collaborate to work towards the planning and implementation of data gathering and reporting to prepare for business-level decision making.

For level 3 organizations, solutions that do not support these goals or require support levels greater than the goals should be re-architected against the gaps they create.

Policy and Environment Usage

OMCM level 4 to 5 activities should focus on reaching a state where the business drives what policies are established and how the environment is used (that is, real business value management). You should analyze your N1 Grid solutions against the need to use N1 Grid efficiency as a tool for strategic differentiation (for example, releasing more revenue-generating services than your competition by using mobility) to achieve production environment stability and to reduce new service time to market.

At this stage, the IT organization should be:

  • Focused on process improvement and on adding quantifiable value to the business

  • Providing data to the management infrastructure for use in modifying processes to gain efficiencies

For level 4 organizations, solutions that do not support these goals or require support levels greater than these goals should be re-architected against the gaps they create.



Buliding N1 Grid Solutions Preparing, Architecting, and Implementing Service-Centric Data Centers
Buliding N1 Grid Solutions Preparing, Architecting, and Implementing Service-Centric Data Centers
ISBN: N/A
EAN: N/A
Year: 2003
Pages: 144

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net