[SOLR-13898] Non-atomic use of SolrCache get / put

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 8.3
• Fix Version/s: 8.4
• Assignee: Andrzej Bialecki
• Reporter: Andrzej Bialecki
• Created: 05/Nov/19 19:12
• Updated: 21/Jan/20 16:58
• Resolved: 05/Dec/19 18:16

Description

As pointed out by ben.manes in SOLR-13817 Solr code base in many key places uses a similar pattern of non-atomic get / put calls to SolrCache-s. In multi-threaded environment this leads to cache misses and additional unnecessary computations when competing threads both discover a missing value, non-atomically compute it and update the cache.

Some of these places are known performance bottlenecks where efficient caching is especially important, such as SolrIndexSearcher, SolrDocumentFetcher, UninvertedField and join queries .

I propose to add SolrCache.computeIfAbsent(key, mappingFunction) that will atomically retrieve existing values or compute and update the cache. This will require also changing how the SolrCache.get(...) is used in many components.