[LUCENE-1600] Reduce usage of String.intern(), performance is terrible

• Type: Improvement
• Status: Closed
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 2.4, 2.4.1
• Fix Version/s: 2.9

Environment

Windows Server 2003 x64
Hotspot JDK 1.6.0_12 64-bit

• Lucene Fields: New, Patch Available
• Assignee: Michael McCandless
• Reporter: Patrick Eger
• Created: 13/Apr/09 18:49
• Updated: 28/Aug/22 11:59
• Resolved: 13/Apr/09 20:07

Description

I profiled a simple MatchAllDocsQuery() against ~1.5 million documents (8 fields of short text, Field.Store.YES,Field.Index.NOT_ANALYZED_NO_NORMS), then retrieved all documents via searcher.doc(i, fs). String.intern() showed up as a top hotspot (see attached screenshot), so i implemented a small optimization to not intern() for every new Field(), instead forcing the intern in the FieldInfos class and adding a optional "internName" constructor to Field. This reduced execution time for searching and iterating through all documents by 35%. Results were similar for -server and -client.

TRUNK (2.9) w/out patch: matched 1435563 in 8884 ms/search
TRUNK (2.9) w/patch: matched 1435563 in 5786 ms/search

Issue Links

• is related to: LUCENE-1308 Remove String.intern() from Field.java to increase performance and lower contention
• is related to: LUCENE-1607 String.intern() faster alternative
• relates to: LUCENE-2545 improve uses of StringHelper.intern in Field,AbstractField, NumericField etc.