Unlike the Debezium connector for MySQL which consumes the changes from all databases via a single binlog, the SQL Server connector interacts with the CDC schema which is deployed individually for each database.
At SugarCRM, we have a number of SQL Server instances hosting more than a hundred databases each, the changes from which we'd like to capture using the Debezium connector for SQL Server. Having to deploy a hundred connectors per instance may be suboptimal both from the resourcing and the operations standpoints.