msokolov edited a comment on pull request #617: URL: https://github.com/apache/lucene/pull/617#issuecomment-1018061477
Well, it's certainly a nice cleanup. The performance results are confusing at first glance, since we now write the temp file and then copy it, but consider that now when we traverse the vectors while writing the graph, we are accessing a file (that we just wrote, so presumably its pages are still in the file system cache) that densely stores the vectors, where before we were accessing those same vectors through an array of pointers, requiring an extra lookup, so it seems that it's a net cost savings! awesome! -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
