mistercrunch commented on issue #7804: URL: https://github.com/apache/incubator-superset/issues/7804#issuecomment-697144955
Some notes here: - if we're not caching the result set, this means every page switch means a database scan - pagination to be deterministic assumes ordering, meaning that in @bolkedebruin 's example of 25M rows with 100 row pages, skimming through 10 pages of 100 row each means scanning / sorting 25M row for each page - this seems like potentially a lot work for the database to serve little to the user Ultimately we could have "advanced pagination" options that let the user choose between server and client-side pagination options. I think that when people say "I want to browse through millions of rows" it's pointing to other problems. It very rarely makes sense for anyone to comb through millions of individual rows. ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
