dantuzi opened a new pull request, #12246: URL: https://github.com/apache/lucene/pull/12246
### Description This is a quick mitigation to avoid unexpected behavior if multiple queries are executed concurrently. This commit wants to address some comments received in PR [12169](https://github.com/apache/lucene/pull/12169). As @jimczi pointed out, the `OnHeapHnswGraph` is not thread-safe so we can avoid concurrent access to that object synchronizing the requests. Indeed, the graph has been reset by the searcher that calls the method `seek` [before each search](https://github.com/apache/lucene/blob/776149f0f6964bbc72ad2d292d1bfe770f82ba45/lucene/core/src/java/org/apache/lucene/util/hnsw/HnswGraphSearcher.java#L255). Talking about performance, we will attach the result of some benchmarks we executed. @zhaih noticed that the number of nodes that can be visited during the HnswGraph search was limited so I increased that limit to the max integer. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
