diqiu50 opened a new issue, #13697: URL: https://github.com/apache/gravitino/issues/13697
### What would you like to be improved? On every metadata refresh (`gravitino.metadata.refresh-interval-seconds`, 10s by default), the Trino connector calls `listCatalogs()` and then `loadCatalog()` for each catalog, even when nothing has changed. That is 1 + N requests per metalake per cycle, and each one is written to the audit log. With 15 catalogs, an idle cluster writes about 140,000 catalog read records a day, and they look the same as reads by a real user. ### How should we improve? Fetch all catalogs with one `listCatalogsInfo()` call per cycle. If that listing fails, fall back to loading catalogs one by one, so that one catalog that can't be resolved doesn't block the rest. -- 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]
