Problem: My database has 2,000 tables and my Reverse Engineer is very slow. How long does erwin take to reverse engineer a database?
The time for the Reverse Engineer (RE) depends upon the number of objects and properties that erwin must create, how powerful a machine you are using, and which options you are selecting in the RE-set options. These are the options that most likely can degrade performance when doing a Reverse Engineer:
1. Row Count
2. Average Width
3. Percent Null
4. Infer Relationships
Infer relationships can be very slow because erwin has to do many calculations in memory and that is based on the foreign keys and the indexes.
Please deselect those options in the Reverse Engineer 'Option Set' and the slow performance will most likely subside.
Here are some additional strategies that could also help speed up an erwin reverse engineer:
1. RE the specific schema's as separate models.
During the RE, specify one schema owner and RE that schema first. Then RE the second schema, and so on as separate models. You can later combine the models together as one large model if you want to do so using complete compare.
2. Try checking off System Objects and Use DBA Views if they are checked on.
3. If you are reverse engineering a database that supports tablespaces, you can press the IN button and type one tablespace or just a few tablespaces. Reverse Engineering one or just a few tablespaces at a time could also quicken the reverse engineer.