10925 Rate this article:
No rating

Configuration Management and Enterprise GIS

Mark Alonzo

With the Esri Federal GIS conference right around the corner, it seems like a good time to look forward and examine what government GIS leaders will be focusing on this year, and what key pieces of information they will be looking to take away from the conference.   For one thing, federal GIS professionals will be looking to doing more with less this year and searching for ways to increase efficiency.

Certainly one of the most advantageous ways to do this could be to move GIS and image analytics services to existing enterprise infrastructures.  With this technique services can be consolidated at the enterprise level.  This centralization will not only reduce the overhead associated with managing the IT infrastructure, but it also allows for another key benefit - ease of executing configuration management on  an organization’s vetted workflows and products.

First off, a quick summary of configuration management (CM), Wikipedia informs us that “CM is the practice of handling changes systematically so that a system maintains its integrity over time”.  Applying this to image analytics and GIS workflows, the ‘configuration’ refers not so much to the hardware and software components of a system, but rather to the data sources, metadata, workflow, and processing settings.

For complex image analytics and GIS workflows a GIS professional might have hundreds of possible adjustments or settings at their disposal.  If different users make use of differing settings, the results can vary widely. We’ve all heard the saying… “garbage in…garbage out…”.

2_7_Classification

This flexibility in configuring the workflow and processing settings should not be viewed negatively, as it is absolutely required. Subject matter experts are fully aware of the effects they are introducing by performing these adjustments, and they require this flexibility to ensure the best product possible is obtained.  However, in many cases, once a workflow has been finalized by subject matter experts, the settings can remain unchanged for future executions with different source data.

Within a configuration management process, once the settings have been formally accepted by an organization, they are then entered into the organization’s configuration management system and deployed to the enterprise.  It is expected that things may change with time, and that is partially why the practice of configuration management exists - to properly vet, test, and control the release of these evolutionary changes.

It is at this deployment phase of the configuration management cycle, where the ‘enterprise’ aspect of enterprise image analytics and GIS can shows its value.   The vetted workflow, with the CM approved settings, can be essentially released simultaneously to all enterprise users, by moving the service and approved settings to the enterprise infrastructure.    ENVI Services Engine can streamline the transition from the desktop, where the services are created and refined, to the enterprise, where the approved configuration managed services are implemented.    In terms of efficiencies - once deployed at the enterprise level, all users, even those who are not subject matter experts, can perform complex, proven image and GIS analytics to achieve accurate results. As the service is deployed at one centralized location, it does not need to be pushed out separately to all the desktops within the organization.  This centralization greatly eases the process of implementing changes to the approved configuration baseline.

So back to the Federal GIS conference.  How do you see the practice of configuration management being used for image analytics and GIS in your organization? Are you excited to learn about configuration management or enterprise implementation at this years’ conference?

Please login or register to post comments.