topic Re: How To Best Manage Data in SDE? in Data Management Questions

How To Best Manage Data in SDE?

SFGISProgram — Tue, 16 Sep 2014 15:34:20 GMT

Hi.... What's the best way to organize our primary enterprise Geodatabase? We want to break it down into separate, more manageable data categories (Ex. Utilities, Operations, Environmental, etc.). Any recommendations as to doing this at the Geodatabase, Schema, or Feature Dataset level? What are the pros/cons to each scenario? Thank you.

Re: How To Best Manage Data in SDE?

VinceAngelo — Tue, 16 Sep 2014 15:46:20 GMT

That's a really broad question, which has multiple possible answers, depending on implementation details that haven't been provided...

What version of Oracle are you using (exact release, as in 11.2.0.3.1)?
What version of ArcGIS are you using? What patches are installed?
How many different data sources do you have? How many different tables in each category? How may rows in each table? How would you categorize them in volatility (read-only, read-mostly, read-write, write-mostly)? What is the total expected data volume (in gigabytes)?
How many concurrent GIS users do you have? Are different departments involved? What kind of isolation is required?

- V

Re: How To Best Manage Data in SDE?

SFGISProgram — Tue, 16 Sep 2014 17:54:59 GMT

Thanks for your response Vince... To clarify, here are some more details for you:

Oracle 11G v2
ArcGIS 10.2 v2
Data Sources: Basemap, Environmental, Operations, Planning, Utilities
Feature Classes: Basemap (30), Environmental (10), Operations (30), Planning (15), Utilities (30)
Table Rows: On Avg. - Basemap (30,000 - 40,000), Environmental (5,000-10,0000), Operations (100-500), Planning (500-1000), Utilities (10,000-20,000)
Volatility: All read-write (based on category)
Total Volume: 2.17GB
Concurrent Users: 1 GIS Admin, 2 GIS Analysts, 1 GIS/Utilities Specialist, 100+ viewers (online), 10 mobile
Departments: Engineering (primary), Operations, Planning

Hope this clarifies a bit for you... Sorry for the generalization earlier.

Thank you,

Jason