Github user ravipesala commented on the issue: https://github.com/apache/incubator-carbondata/pull/624 Yes, it fits in OS cache, that is the same case with parquet as well so it would be fare comparison. What I am suggesting is now we run the queries of parquet and carbon together. So there is a possibility of one engine jvm memory consumption can effect the other engine. It would be better to run all the queries of one engine and flush the GC and then start running other engine queries.
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---