More and more frequently, database-as-a-service (DAAS) applications are emerging on the web. In this paper, we address the problem of evaluating a heavy load of subscribed queries in the DAAS model, where the traditional XML data management strategy leads to many drawbacks, such as overload of the service provider and the waste of bandwidth when a large number of service queries are executed on the service provider due to huge XML message sizes. We propose a new framework to efficiently process multi-queries among the cooperative service clients. System maintenance issues are also discussed. The experimental results demonstrate that the proposed approach substantially saves more bandwidth than traditional ones.