In the FME 2016, FeatureReader is added a parameter called "Max Feature to Read", This is very useful when test reader result, because you can set a smaller value (eg.10), so don't have to read all of the data can be test reader result.
But such as in the Test directory with a.shp,b.shp,c.shp, specified in the Dataset to [Test *. SHP], this will only be able to read the a.shp's before 10 features, and need a lot of time in each file read before 10 features.
So I hope as well as individual Reader, to add a "Max Feature to Read Per Feature Type" parameters to the FeatureReader.
The option is there now as it seems but not really working in general. I tested it with FME 2017 for SQLite non spatial.
I set Max features to read per feature type: 10 and ran it against a very large SQLite DB.
It took 15 min to read and presented me with 10 features of the first found table.
Quite disappointing result ...
Please check out my comments on this post for a suggested workaround: https://community.safe.com/s/question/0D54Q000080hEo2SAE/max-number-of-features-per-feature-type-for-a-featurereader