A company needs to deploy Amazon Redshift in a multi-AZ setup to ensure higher availability and fault tolerance for their data warehouse workloads. They also want to separate compute and storage scaling to handle varying query loads and data growth. Which node type should they use for this setup?