vvivekiyer commented on issue #9061: URL: https://github.com/apache/pinot/issues/9061#issuecomment-1185791898
Let’s say we create a forward index file where raw values are stored (instead of dictIds). In addition, we also create an inverted index file (obviously with a dictionary). There could be a performance improvement (albeit not large) when values are looked up from the forward index file (additional hop of forwardIndex -> DictId -> Value is eliminated. However, the storage utilized might be large. So, we are trading space for time. Based on the [attached table](https://docs.google.com/document/d/1SBXsV1lH1TRxJYI-Ulevi3nrXEAovL4j5DNE7Pv0i-A/edit#) which outlines the space bloat for supporting inverted index for raw columns, we believe the benefits of supporting inverted index for raw columns does not make sense. -- 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]
