A reminder about this message from last week. Please be thinking about your need for a database to support your work.
We'll start discussing at this week's GHE meeting on Wednesday.
The GHE is undertaking a review of our data management infrastructure over the course of the next few months based on the recommendations included in the Esri EEAP report. There are a couple of things that we are starting to investigate more closely and we're looking for preliminary input.
2.1.6 Use File GDBs for Read-Only Publication Geodatabases
Esri recommends the GHE team use file Geodatabases to store and disseminate their read-only geo data for both internal and public consumption. This will simplify data management and system architecture (e.g. less server components) and optimize performance (e.g. local file storage is usually faster than DBMS).
This is a direction that we'd like to pursue. Data being shared can either be stored on G&M servers (as part of their ProjectOne/CTS workflows) and/or on \\ghearc.lcds.loc.gov.
We can start doing this at any time, but perhaps there's a pilot project to get things started?
2.1.7 Use Enterprise GDBs for Web Editing Geodatabases
Esri recommends the GHE team continue to use a DBMS in the GHE system architecture to host the enterprise Geodatabase and exclusively provide web editing capabilities, like mobile collection, browser-based web editing sessions, etc. The DBMS can be Postgres, SQL Server, IBM DB2, Oracle, Informix.
After a few years of the GHE we've realized that we don't really need a database for the majority of "static" content that we have. However, there is likely a need for a "dynamic" database infrastructure down the road. The GHE has a plan to migrate from Oracle to PostGres as the "database" component of the GHE for dynamic data, but before that we need to decide if anything that is in the current GHE databases (LRP, PUB, RAP) has "value." "Value" brings up a few questions:
1. Is the copy in Oracle the only copy available?
2. If not, does if have to stay in Oracle?
3. Are there any active maps/applications that are built directly on the content in Oracle?
4. If the Oracle database went away would it affect any active maps/apps?
5. Can the existing Oracle content be packaged and shared via a file geodatabase?
6. If all the current content in the Oracle databases can be accounted for by the above, can we delete the content from the Oracle databases?
7. If all the current content in the Oracle databases can be accounted for by the above, is there any need to keep the Oracle databases at this time?
Please start thinking about the above issues. We'll start to pursue this work more formally over the next couple of months, but these are the main questions we'll be asking at the start.
IT Project Manager
Geospatial Hosting Environment project
Library of Congress
[log in to unmask]