A data engineer wants to implement Redshift Workload Management (WLM) to prioritize short-running queries over long-running analytical queries. The engineer wants to allocate system resources dynamically based on query complexity and avoid manual configuration of memory and concurrency settings. Which WLM mode should the engineer choose?