You have a Fabric workspace named Workspace1 that contains multiple semantic models, including one named Model1. Model1 is updated via an XMLA endpoint. What should you do to increase the speed of these write operations?