Our initial results in this study are obtained from a combination of high-level abstraction and practical implementation from prototype Linux-based clusters of 64 nodes, wide-area planet lab deployment and ORBIT wireless testbed. These results suggest that (1) our CLB/AR programming abstraction provides convenient programming methodologies; (2) Application queries can be post to the system with guarantees in terms of number of messages and number of peers involved; (3) The decentralized aggregation service guarantees that all data that matches a query will be located and aggregated. (4) The overheads at each layer of Meteor Stack were measured to verify the scalability of the Meteor infrastructure in terms of network size.