Group by Similar Retention
The Group by Similar Retention view groups objects by similarity of their retention paths. For each object set, the view shows two shortest paths to roots. Note that from the whole variety of shortest paths, dotMemory selects only two paths that are most different from each other.
The Group by Similar Retention view allows you to distinguish where the objects of the same type logically belong to. For example, it's obvious that a number of int variables in an array will have the same retention path which is different to variables of any other int array.
You can also use Group by Similar Retention to identify why an object is still in memory. For example, it may be useful when investigating potential memory leaks caused by objects retained on more than one path (see Group by Dominators for details).
The list of objects for which graph is built consists of the following columns:
The number of objects with the same retention path.
The overall size of objects in bytes.
You can filter out objects that are of no interest for your analysis.
To narrow the list
- Start typing the desired type name in the Filter field.
dotMemory will exclude all instances that don't match the pattern.
You can make your search more efficient by using the following tips:
Use CamelHumps. E.g.
fowill return objects of both
Use special symbols, like wildcards and others. The full list is shown in the table below.
Leave only arrays
Leave only arrays of the specified or higher (if brackets are not closed) dimension
Exclude arrays from the result
Generic type arguments
Leave only types with generic type arguments
Leave only objects with the specified number of generic type arguments
Exclude generic type arguments from the search scope
Selecting objects for further analysis
In the Group by Similar Retention view, you can select the following subjects for further analysis: