DCjanus commented on issue #1658: URL: https://github.com/apache/kvrocks/issues/1658#issuecomment-2887477667
After a deeper look, I realized that RocksDB’s `GetPropertiesOfAllTables` is not a lightweight or lazy operation—it eagerly loads all table properties into memory, which can be quite costly for large instances with many SST files. This makes the SST sampling approach less practical than I initially thought. Given these trade-offs, I’m now considering the multiple-cursor solution that was suggested earlier in this issue. This approach seems much more efficient and scalable, especially for large datasets, and avoids the overhead of scanning all SST properties. Thank you for highlighting this alternative and for your valuable input—it’s been very helpful in guiding the direction of this optimization! -- 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]
