[SPARK-34309] Use Caffeine instead of Guava Cache

• Type: Improvement
• Status: In Progress
• Priority: Minor
• Resolution: Unresolved
• Affects Version/s: 3.2.0
• Component/s: Spark Core, SQL
• Assignee: Unassigned
• Reporter: Yang Jie
• Created: 01/Feb/21 08:11
• Updated: 12/Dec/22 18:10

Description

Caffeine is a high performance, near optimal caching library based on Java 8, it is used in a similar way to guava cache, but with better performance. The comparison results are as follow are on the caffeine benchmarks

At the same time, caffeine has been used in some open source projects like Cassandra, Hbase, Neo4j, Druid, Spring and so on.

 

Issue Links

• links to: [Github] Pull Request #31517 (LuciferYang)
• links to: [Github] Pull Request #31517 (LuciferYang)
• links to: [Github] Pull Request #33772 (sarutak)
• links to: [Github] Pull Request #33772 (sarutak)
• links to: [Github] Pull Request #33784 (LuciferYang)
• links to: [Github] Pull Request #33784 (LuciferYang)