How do you optimize a slow-running query in Snowflake at Sedgwick? Discuss your approach to micro-partitioning and clustering keys.
