A lakehouse maintenance task aims to improve query performance and reduce storage costs for a Delta table with seven days retention and many small files. Which combination of commands should you run?

A lakehouse maintenance task aims to improve query performance and reduce storage costs for a Delta table with seven days retention and many small files. Which combination of commands should you run?

• A. Run the VACUUM command under Maintenance
• B. Run only the OPTIMIZE command
• C. Run the OPTIMIZE command and then the VACUUM command with a retention policy of seven days
• D. Run a VACUUM and an ANALYZE command

Answer: (C)

When you have many small files, query performance suffers and storage costs stay high because the engine has to open and merge lots of tiny pieces. The key steps are to first consolidate those small files and then remove the older, unnecessary files. Running the optimization step reshapes the data layout by combining small files into larger base files, which speeds up scans and reduces the overhead of reading many tiny files. After that, applying a vacuum with a seven-day retention cleans up files that are no longer needed beyond that retention window. This preserves only the data from the most recent seven days while freeing up storage space from obsolete files and artifacts created during optimization. Running only the vacuum wouldn’t improve performance much if the data remains fragmented with many small files. Running only optimization leaves older, unnecessary files in storage. And ANALYZE doesn’t address file fragmentation or removal of old data; it gathers statistics for query planning rather than housekeeping. So the best approach is to optimize first to improve layout, then vacuum with a seven-day retention to prune outdated files.

When you have many small files, query performance suffers and storage costs stay high because the engine has to open and merge lots of tiny pieces. The key steps are to first consolidate those small files and then remove the older, unnecessary files.

Running the optimization step reshapes the data layout by combining small files into larger base files, which speeds up scans and reduces the overhead of reading many tiny files. After that, applying a vacuum with a seven-day retention cleans up files that are no longer needed beyond that retention window. This preserves only the data from the most recent seven days while freeing up storage space from obsolete files and artifacts created during optimization.

Running only the vacuum wouldn’t improve performance much if the data remains fragmented with many small files. Running only optimization leaves older, unnecessary files in storage. And ANALYZE doesn’t address file fragmentation or removal of old data; it gathers statistics for query planning rather than housekeeping. So the best approach is to optimize first to improve layout, then vacuum with a seven-day retention to prune outdated files.