In the span of 10 years, the amount of data an average-sized city has to manage can shoot up considerably, with its GIS potentially mushrooming from 200 to over 1,000 layers of all kinds. This increase brings with it a number of other challenges, as this information has to be processed and stored before it can eventually be used and viewed by city employees and external stakeholders.
A Geodatabase is usually the go-to storage solution for organizations that have attained a certain level of maturity and now want to centralize their information and make sure their spatial data is shared and distributed as effectively as possible.
Spatial data is a strategic asset that requires regular backup to guard against disasters, system failures and technological problems like data corruption. In some cases, backups can double as an archiving solution, keeping historicity when this type of solution proves too expensive or complex for the organization to manage.
Esri’s replication tools
Some organizations might look at the many tools and strategies available and at their own situation and come to the conclusion that they would be best served by other techniques. That is where FME’s versatility comes in with a powerful solution to automate the Geodatabase backup process.
By automating the backup function of your Geodatabase, you can avoid unfortunate oversights that put you at risk of losing all the changes made during a given period.
What’s more, automation means work can be done at a specific time, preferably at night, to reduce service interruptions and other maintenance constraints during normal operating hours. It also makes for a more efficient environment for the production team, with much lower latency.
Case in point: Granby, a growing municipality with a population of nearly 70,000. The City’s automation plan, which included a database defragmentation strategy, made sure the 150 ArcGIS users (both internal employees and external partners) were always working with purified, up-to-date information.
FME to the rescue
Granby’s strategy was to use a main (parent) FME workspace along with secondary (child) workspaces to automate a series of operations required to create a temporary Geodatabase to be subsequently deployed into production.
The sequence kicks off in the middle of the night, with the first workspace compressing and defragmenting the spatial database, using the ArcPy API from ArcGIS. A second script, which also uses ArcPy, extracts the data into a temporary Geodatabase.
A third workspace uses a technique to rename the production geodatabase and runs a Python script to suspend the web service to which it is connected. This database then becomes the backup database. The fourth step is to rename the temporary geodatabase according to the naming convention for the production environment, which the new database now qualifies as. One last automation step reactivates the web service, reoriented toward the newly created production geodatabase.
This process also automatically restores the original database, in case there is an error somewhere along the way. This way, ArcGIS application and map users can enjoy smooth, uninterrupted service.
For the City of Granby, using FME to automate its backups every night is a highly advantageous approach, which allows ArcGIS users to work in an optimal, efficient environment with no unpleasant service interruptions during the day. In addition, defragmentation gives them the benefit of working with purified information.
In Granby’s case, with a Geodatabase of over 2 GB, this entire process, from creating the backup to deploying the new Geodatabase, could take over 30 minutes—a sizeable service interruption. The sequence we described cuts users off from the web service for a mere few seconds.
In conclusion, Geodatabases are a valuable asset for any organization. An effective regular backup strategy is important, to protect against disasters and potentially keep a history. In many cases, the Esri suite offers the best solution, while other situations call for more personalized solutions, and that’s where FME comes in.
Are you looking to implement your own Geodatabase backup strategy?