[COLLECTIONS-413] Performance problem in DualHashBidiMap

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 3.2.1
• Fix Version/s: 4.0-alpha1, 4.0

Environment

java 1.6.0_24
Ubuntu 11.10

• Assignee: Unassigned
• Reporter: Adrian Nistor
• Created: 21/Jun/12 18:38
• Updated: 09/Nov/14 14:31
• Resolved: 23/Jun/12 12:06

Description

Hi,

I am encountering a performance problem in DualHashBidiMap. It
appears in version 3.2.1 and also in revision 1352574 (21 June 2012).
I attached a test that exposes this problem and a patch that fixes it.
On my machine, for this test, the patch provides a 173X speedup.

To run the test, just do:

$ java Test

The output for the un-patched version is:
Time is 5029

The output for the patched version is:
Time is 29

The attached test shows that, for a "DualHashBidiMap bidi" object, the
following operation is very slow:

bidi.entrySet().removeAll(toRemove);

DualHashBidiMap.entrySet() returns a
"DualHashBidiMap.EntrySet" object, which inherits
removeAll(Collection<?> coll) from "DualHashBidiMap.View".

As the patch shows, the problem is that
"DualHashBidiMap.View.removeAll(Collection<?> coll)" performs
"coll.contains(it.next())" for each element in the View.
"coll.contains(it.next())" can be very slow, e.g., if "coll" is a
list.

The patch avoids this cost by using remove(Object obj) (defined in
"EntrySet<K, V>", "KeySet<K>", and "Values<V>"), which is fast because
it uses only operations on sets.

Is this a bug, or am I misunderstanding something? If so, can you
please confirm that the patch is correct?

Thanks,

Adrian